Arrow Research search

Author name cluster

Zhenguo Li

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

89 papers
2 author rows

Possible papers

89

AAAI Conference 2026 Conference Paper

Empowering Sparse-Input Neural Radiance Fields with Dual-Level Semantic Guidance from Dense Novel Views

  • Yingji Zhong
  • Kaichen Zhou
  • Zhihao Li
  • Lanqing Hong
  • Zhenguo Li
  • Dan Xu

Neural Radiance Fields (NeRF) have shown remarkable capabilities for photorealistic novel view synthesis. One major deficiency of NeRF is that dense inputs are typically required, and the rendering quality will drop drastically given sparse inputs. In this paper, we highlight the effectiveness of rendered semantics from dense novel views, and show that rendered semantics can be treated as a more robust form of augmented data than rendered RGB. Our method enhances NeRF’s performance by incorporating guidance derived from the rendered semantics. The rendered semantic guidance encompasses two levels: the supervision level and the feature level. The supervision-level guidance incorporates a bi-directional verification module that decides the validity of each rendered semantic label, while the feature-level guidance integrates a learnable codebook that encodes semantic-aware information, which is queried by each point via the attention mechanism to obtain semanticrelevant predictions. The overall semantic guidance is embedded into a self-improved pipeline.We also introduce a more challenging sparse-input indoor benchmark, where the number of inputs is limited to as few as 6. Experiments demonstrate the effectiveness of our method and it exhibits superior performance compared to existing approaches.

PDF Details DOI

ICLR Conference 2025 Conference Paper

Accelerating Auto-regressive Text-to-Image Generation with Training-free Speculative Jacobi Decoding

  • Yao Teng
  • Han Shi
  • Xian Liu
  • Xuefei Ning
  • Guohao Dai 0001
  • Yu Wang 0002
  • Zhenguo Li
  • Xihui Liu

The current large auto-regressive models can generate high-quality, high-resolution images, but these models require hundreds or even thousands of steps of next-token prediction during inference, resulting in substantial time consumption. In existing studies, Jacobi decoding, an iterative parallel decoding algorithm, has been used to accelerate the auto-regressive generation and can be executed without training. However, the Jacobi decoding relies on a deterministic criterion to determine the convergence of iterations. Thus, it works for greedy decoding but is incompatible with sampling-based decoding which is crucial for visual quality and diversity in the current auto-regressive text-to-image generation. In this paper, we propose a training-free probabilistic parallel decoding algorithm, Speculative Jacobi Decoding (SJD), to accelerate auto-regressive text-to-image generation. By introducing a probabilistic convergence criterion, our SJD accelerates the inference of auto-regressive text-to-image generation while maintaining the randomness in sampling-based token decoding and allowing the model to generate diverse images. Specifically, SJD facilitates the model to predict multiple tokens at each step and accepts tokens based on the probabilistic criterion, enabling the model to generate images with fewer steps than the conventional next-token-prediction paradigm. We also investigate the token initialization strategies that leverage the spatial locality of visual data to further improve the acceleration ratio under specific scenarios. We conduct experiments for our proposed SJD on multiple auto-regressive text-to-image generation models, showing the effectiveness of model acceleration without sacrificing the visual quality. The code of our work is available here: https://github.com/tyshiwo1/Accelerating-T2I-AR-with-SJD/.

Details

TMLR Journal 2025 Journal Article

AlgoFormer: An Efficient Transformer Framework with Algorithmic Structures

  • Yihang Gao
  • Chuanyang Zheng
  • Enze Xie
  • Han Shi
  • Tianyang Hu
  • Yu Li
  • Michael Ng
  • Zhenguo Li

Besides natural language processing, transformers exhibit extraordinary performance in solving broader applications, including scientific computing and computer vision. Previous works try to explain this from the expressive power and capability perspectives that standard transformers are capable of performing some algorithms. To empower transformers with algorithmic capabilities and motivated by the recently proposed looped transformer (Yang et al., 2024; Giannou et al., 2023), we design a novel transformer framework, dubbed Algorithm Transformer (abbreviated as AlgoFormer). We provide an insight that efficient transformer architectures can be designed by leveraging prior knowledge of tasks and the underlying structure of potential algorithms. Compared with the standard transformer and vanilla looped transformer, the proposed AlgoFormer can perform efficiently in algorithm representation in some specific tasks. In particular, inspired by the structure of human-designed learning algorithms, our transformer framework consists of a pre-transformer that is responsible for task preprocessing, a looped transformer for iterative optimization algorithms, and a post-transformer for producing the desired results after post-processing. We provide theoretical evidence of the expressive power of the AlgoFormer in solving some challenging problems, mirroring human-designed algorithms. Furthermore, some theoretical and empirical results are presented to show that the designed transformer has the potential to perform algorithm representation and learning. Experimental results demonstrate the empirical superiority of the proposed transformer in that it outperforms the standard transformer and vanilla looped transformer in some specific tasks. An extensive experiment on real language tasks (e.g., neural machine translation of German and English, and text classification) further validates the expressiveness and effectiveness of AlgoFormer.

PDF Details

ICLR Conference 2025 Conference Paper

Beyond Autoregression: Discrete Diffusion for Complex Reasoning and Planning

  • Jiacheng Ye
  • Jiahui Gao
  • Shansan Gong
  • Lin Zheng
  • Xin Jiang
  • Zhenguo Li
  • Lingpeng Kong

Autoregressive language models, despite their impressive capabilities, struggle with complex reasoning and long-term planning tasks. We introduce discrete diffusion models as a novel solution to these challenges. Through the lens of subgoal imbalance, we demonstrate how diffusion models effectively learn difficult subgoals that elude autoregressive approaches. We propose Multi-Granularity Diffusion Modeling (MGDM), which prioritizes subgoals based on difficulty during learning. On complex tasks like Countdown, Sudoku, and Boolean Satisfiability Problems, MGDM significantly outperforms autoregressive models without using search techniques. For instance, MGDM achieves 91.5\% and 100\% accuracy on Countdown and Sudoku, respectively, compared to 45.8\% and 20.7\% for autoregressive models. Our work highlights the potential of diffusion-based approaches in advancing AI capabilities for sophisticated language understanding and problem-solving tasks. All associated codes are available at \href{https://github.com/HKUNLP/diffusion-vs-ar}{https://github.com/HKUNLP/diffusion-vs-ar}.

Details

ICLR Conference 2025 Conference Paper

CARTS: Advancing Neural Theorem Proving with Diversified Tactic Calibration and Bias-Resistant Tree Search

  • Xiao-Wen Yang
  • Zhi Zhou
  • Haiming Wang
  • Aoxue Li
  • Wen-Da Wei
  • Hui Jin
  • Zhenguo Li
  • Yu-Feng Li

Recent advancements in neural theorem proving integrate large language models with tree search algorithms like Monte Carlo Tree Search (MCTS), where the language model suggests tactics and the tree search finds the complete proof path. However, many tactics proposed by the language model converge to semantically or strategically similar, reducing diversity and increasing search costs by expanding redundant proof paths. This issue exacerbates as computation scales and more tactics are explored per state. Furthermore, the trained value function suffers from false negatives, label imbalance, and domain gaps due to biased data construction. To address these challenges, we propose CARTS (diversified tactic CAlibration and bias-Resistant Tree Search), which balances tactic diversity and importance while calibrating model confidence. CARTS also introduce preference modeling and an adjustment term related to the ratio of valid tactics to improve the bias-resistance of the value function. Experimental results demonstrate that CARTS consistently outperforms previous methods achieving a pass@l rate of 49.6\% on the miniF2F-test benchmark. Further analysis confirms that CARTS improves tactic diversity and leads to a more balanced tree search.

Details

ICLR Conference 2025 Conference Paper

Forewarned is Forearmed: Harnessing LLMs for Data Synthesis via Failure-induced Exploration

  • Qintong Li
  • Jiahui Gao
  • Sheng Wang
  • Renjie Pi
  • Xueliang Zhao
  • Chuan Wu
  • Xin Jiang
  • Zhenguo Li

Large language models (LLMs) have significantly benefited from training on diverse, high-quality task-specific data, leading to impressive performance across a range of downstream applications. Current methods often rely on human-annotated data or predefined task templates to direct powerful LLMs in synthesizing task-relevant data for effective model training. However, this dependence on manually designed components may constrain the scope of generated data, potentially overlooking critical edge cases or novel scenarios that could challenge the model. In this paper, we present a novel approach, ReverseGen, designed to automatically generate effective training samples that expose the weaknesses of LLMs. Specifically, we introduce a dedicated proposer trained to produce queries that lead target models to generate unsatisfactory responses. These failure-inducing queries are then used to construct training data, helping to address the models' shortcomings and improve overall performance. Our approach is flexible and can be applied to models of various scales (3B, 7B, and 8B). We evaluate ReverseGen on three key applications—safety, honesty, and math—demonstrating that our generated data is both highly effective and diverse. Models fine-tuned with ReverseGen-generated data consistently outperform those trained on human-annotated or general model-generated data, offering a new perspective on data synthesis for task-specific LLM enhancement.

Details

NeurIPS Conference 2025 Conference Paper

FUDOKI: Discrete Flow-based Unified Understanding and Generation via Kinetic-Optimal Velocities

  • Jin Wang
  • Yao Lai
  • Aoxue Li
  • Shifeng Zhang
  • Jiacheng Sun
  • Ning Kang
  • Chengyue Wu
  • Zhenguo Li

The rapid progress of large language models (LLMs) has catalyzed the emergence of multimodal large language models (MLLMs) that unify visual understanding and image generation within a single framework. However, most existing MLLMs rely on autoregressive (AR) architectures, which impose inherent limitations on future development, such as the raster-scan order in image generation and restricted reasoning abilities in causal context modeling. In this work, we challenge the dominance of AR-based approaches by introducing FUDOKI, a unified multimodal model purely based on discrete flow matching, as an alternative to conventional AR paradigms. By leveraging metric-induced probability paths with kinetic optimal velocities, our framework goes beyond the previous masking-based corruption process, enabling iterative refinement with self-correction capability and richer bidirectional context integration during generation. To mitigate the high cost of training from scratch, we initialize FUDOKI from pre-trained AR-based MLLMs and adaptively transition to the discrete flow matching paradigm. Experimental results show that FUDOKI achieves performance comparable to state-of-the-art AR-based MLLMs across both visual understanding and image generation tasks, highlighting its potential as a foundation for next-generation unified multimodal models. Furthermore, we show that applying test-time scaling techniques to FUDOKI yields significant performance gains, further underscoring its promise for future enhancement through reinforcement learning.

PDF Details

ICLR Conference 2025 Conference Paper

G-LLaVA: Solving Geometric Problem with Multi-Modal Large Language Model

  • Jiahui Gao
  • Renjie Pi
  • Jipeng Zhang
  • Jiacheng Ye
  • Wanjun Zhong
  • Yufei Wang 0005
  • Lanqing Hong
  • Jianhua Han

Large language models (LLMs) have shown remarkable proficiency in human-level reasoning and generation capabilities, which encourages extensive research on their application in mathematical problem solving. However, current work has been largely focused on text-based mathematical problems, with limited investigation in problems involving multi-modal geometric information. Addressing this gap, we aim to enable LLMs to solve geometric problems by understanding image input. We first identify the limitations of current Multimodal Large Language Models (MLLMs) in this area: they struggle to accurately comprehend basic geometric elements and their relationships. To address these challenges, we leverage the inherent attribute of logical structure compactness in geometric figures, utilizing text-only Large Language Models (LLMs) to curate a comprehensive multimodal geometry dataset. This dataset, named Geo170k, contains more than 170K geometric image-caption and question-answer pairs. Utilizing the Geo170k dataset, we introduce G-LLaVA, a model that demonstrates exceptional performance in solving geometric problems. It significantly outperforms GPT4-V on the geometry task of MathVista benchmark with only 7B parameters.

Details

ICLR Conference 2025 Conference Paper

Implicit Search via Discrete Diffusion: A Study on Chess

  • Jiacheng Ye
  • Zhenyu Wu
  • Jiahui Gao
  • Zhiyong Wu 0003
  • Xin Jiang 0002
  • Zhenguo Li
  • Lingpeng Kong

In the post-AlphaGo era, there has been a renewed interest in search techniques such as Monte Carlo Tree Search (MCTS), particularly in their application to Large Language Models (LLMs). This renewed attention is driven by the recognition that current next-token prediction models often lack the ability for long-term planning. Is it possible to instill search-like abilities within the models to enhance their planning abilities without relying on explicit search? We propose DiffuSearch , a model that does \textit{implicit search} by looking into the future world via discrete diffusion modeling. We instantiate DiffuSearch on a classical board game, Chess, where explicit search is known to be essential. Through extensive controlled experiments, we show DiffuSearch outperforms both the searchless and explicit search-enhanced policies. Specifically, DiffuSearch outperforms the one-step policy by 19.2\% and the MCTS-enhanced policy by 14\% on action accuracy. Furthermore, DiffuSearch demonstrates a notable 30\% enhancement in puzzle-solving abilities compared to explicit search-based policies, along with a significant 540 Elo increase in game-playing strength assessment. These results indicate that implicit search via discrete diffusion is a viable alternative to explicit search over a one-step policy. All codes are publicly available at \href{https://github.com/HKUNLP/DiffuSearch}{https://github.com/HKUNLP/DiffuSearch}.

Details

ICLR Conference 2025 Conference Paper

Improved Diffusion-based Generative Model with Better Adversarial Robustness

  • Zekun Wang 0001
  • Mingyang Yi
  • Shuchen Xue
  • Zhenguo Li
  • Ming Liu 0004
  • Bing Qin 0001
  • Zhiming Ma

Diffusion Probabilistic Models (DPMs) have achieved significant success in generative tasks. However, their training and sampling processes suffer from the issue of distribution mismatch. During the denoising process, the input data distributions differ between the training and inference stages, potentially leading to inaccurate data generation. To obviate this, we analyze the training objective of DPMs and theoretically demonstrate that this mismatch can be alleviated through Distributionally Robust Optimization (DRO), which is equivalent to performing robustness-driven Adversarial Training (AT) on DPMs. Furthermore, for the recently proposed Consistency Model (CM), which distills the inference process of the DPM, we prove that its training objective also encounters the mismatch issue. Fortunately, this issue can be mitigated by AT as well. Based on these insights, we propose to conduct efficient AT on both DPM and CM. Finally, extensive empirical studies validate the effectiveness of AT in diffusion-based models. The code is available at https://github.com/kugwzk/AT_Diff.

Details

ICLR Conference 2025 Conference Paper

Jailbreaking as a Reward Misspecification Problem

  • Zhihui Xie 0002
  • Jiahui Gao
  • Lei Li 0039
  • Zhenguo Li
  • Qi Liu 0049
  • Lingpeng Kong

The widespread adoption of large language models (LLMs) has raised concerns about their safety and reliability, particularly regarding their vulnerability to adversarial attacks. In this paper, we propose a new perspective that attributes this vulnerability to reward misspecification during the alignment process. This misspecification occurs when the reward function fails to accurately capture the intended behavior, leading to misaligned model outputs. We introduce a metric ReGap to quantify the extent of reward misspecification and demonstrate its effectiveness and robustness in detecting harmful backdoor prompts. Building upon these insights, we present ReMiss, a system for automated red teaming that generates adversarial prompts in a reward-misspecified space. ReMiss achieves state-of-the-art attack success rates on the AdvBench benchmark against various target aligned LLMs while preserving the human readability of the generated prompts. Furthermore, these attacks on open-source models demonstrate high transferability to closed-source models like GPT-4o and out-of-distribution tasks from HarmBench. Detailed analysis highlights the unique advantages of the proposed reward misspecification objective compared to previous methods, offering new insights for improving LLM safety and robustness.

Details

NeurIPS Conference 2025 Conference Paper

Masked Diffusion Models as Energy Minimization

  • Sitong Chen
  • Shen Nie
  • Jiacheng Sun
  • Zijin Feng
  • Zhenguo Li
  • Ji-Rong Wen
  • Chongxuan Li

We present a systematic theoretical framework that interprets masked diffusion models (MDMs) as solutions to energy minimization problems in discrete optimal transport. Specifically, we prove that three distinct energy formulations—kinetic, conditional kinetic, and geodesic energy—are mathematically equivalent under the structure of MDMs, and that MDMs minimize all three when the mask schedule satisfies a closed-form optimality condition. This unification not only clarifies the theoretical foundations of MDMs, but also motivates practical improvements in sampling. By parameterizing interpolation schedules via Beta distributions, we reduce the schedule design space to a tractable 2D search, enabling efficient post-training tuning without model modification. Experiments on synthetic and real-world benchmarks demonstrate that our energy-inspired schedules outperform hand-crafted baselines, particularly in low-step sampling settings.

PDF Details

NeurIPS Conference 2025 Conference Paper

Multi-Objective One-Shot Pruning for Large Language Models

  • Weiyu Chen
  • Hansi Yang
  • Yunhao Gou
  • Han Shi
  • Enliang Hu
  • Zhenguo Li
  • James Kwok

Large Language Models (LLMs) have demonstrated remarkable capabilities across various tasks but require substantial computational resources, limiting their deployment in resource-constrained environments. While one-shot pruning methods can reduce model size without expensive retraining, they typically optimize for single objectives, ignoring LLMs' multi-faceted applications. We introduce Multi-Objective One-Shot Pruning (MOSP), which formulates LLM pruning as a multi-objective optimization problem. MOSP efficiently generates a Pareto set of pruned models representing different capability trade-offs, allowing users to select solutions aligned with their preferences. The proposed approach identifies share core support while enabling specialized support. Experiments across various LLMs and sparsity levels demonstrate MOSP's superior performance in navigating multi-objective trade-offs compared to baseline methods.

PDF Details

ICML Conference 2025 Conference Paper

ProofAug: Efficient Neural Theorem Proving via Fine-grained Proof Structure Analysis

  • Haoxiong Liu
  • Jiacheng Sun
  • Zhenguo Li
  • Andrew Chi-Chih Yao

The synergy between deep learning models and traditional automation tools, such as built-in tactics of the proof assistant and off-the-shelf automated theorem provers, plays a crucial role in developing robust and efficient neural theorem provers (NTPs). However, for proof synthesis with LLMs, previous work applies automation tools either only when explicitly invoked by the model or at a single granularity level, failing to fully exploit their power. To solve this issue, we propose ProofAug, a procedure that equips LLMs with automation methods at various granularities through fine-grained structure analysis of model-generated proof proposals. ProofAug also serves as a versatile plug-and-play module that seamlessly integrates with any tree-search algorithm, enabling our construction of an efficient recursive proving (ERP) module to further enhance performance. The superiority of our method is validated on the miniF2F benchmark using the open-source deepseek-math-7b-base model and the Isabelle proof assistant. Notably, by additionally employing a mixed prompting strategy, we achieve a cumulative pass rate of 66. 0% after curation of the dataset (61. 9% for the original version) with 2100 queries to the model per problem (In contrast, the previous SOTA in Isabelle, Subgoal-XL, only achieves 56. 1% using 16384 queries per problem). We also implement a Lean 4 version of ProofAug that can improve the pass@1 performance of Kimina-Prover-Preview-Distill-1. 5B from 44. 3% to 50. 4% on miniF2F-test. Our code is available at https: //github. com/haoxiongliu/ProofAug.

Details

ICML Conference 2025 Conference Paper

SepLLM: Accelerate Large Language Models by Compressing One Segment into One Separator

  • Guoxuan Chen
  • Han Shi
  • Jiawei Li
  • Yihang Gao
  • Xiaozhe Ren
  • Yimeng Chen
  • Xin Jiang
  • Zhenguo Li

Large Language Models (LLMs) have exhibited exceptional performance across a spectrum of natural language processing tasks. However, their substantial sizes pose considerable challenges, particularly in computational demands and inference speed, due to their quadratic complexity. In this work, we have identified a key pattern: certain seemingly meaningless separator tokens (i. e. , punctuations) contribute disproportionately to attention scores compared to semantically meaningful tokens. This observation suggests that information of the segments between these separator tokens can be effectively condensed into the separator tokens themselves without significant information loss. Guided by this insight, we introduce SepLLM, a plug-and-play framework that accelerates inference by compressing these segments and eliminating redundant tokens. Additionally, we implement efficient kernels for training acceleration. Experimental results across training-free, training-from-scratch, and post-training settings demonstrate SepLLM’s effectiveness. Notably, using the Llama-3-8B backbone, SepLLM achieves over 50% reduction in KV cache on the GSM8K-CoT benchmark while maintaining comparable performance. Furthermore, in streaming settings, SepLLM effectively processes sequences of up to 4 million tokens or more while maintaining consistent language modeling capabilities.

Details

NeurIPS Conference 2025 Conference Paper

Speculative Jacobi-Denoising Decoding for Accelerating Autoregressive Text-to-image Generation

  • Yao Teng
  • Fu-Yun Wang
  • Xian Liu
  • Zhekai Chen
  • Han Shi
  • Yu Wang
  • Zhenguo Li
  • Weiyang Liu

As a new paradigm of visual content generation, autoregressive text-to-image models suffer from slow inference due to their sequential token-by-token decoding process, often requiring thousands of model forward passes to generate a single image. To address this inefficiency, we propose Speculative Jacobi-Denoising Decoding (SJD2), a framework that incorporates the denoising process into Jacobi iterations to enable parallel token generation in autoregressive models. Our method introduces a next-clean-token prediction paradigm that enables the pre-trained autoregressive models to accept noise-perturbed token embeddings and predict the next clean tokens through low-cost fine-tuning. This denoising paradigm guides the model towards more stable Jacobi trajectories. During inference, our method initializes token sequences with Gaussian noise and performs iterative next-clean-token-prediction in the embedding space. We employ a probabilistic criterion to verify and accept multiple tokens in parallel, and refine the unaccepted tokens for the next iteration with the denoising trajectory. Experiments show that our method can accelerate generation by reducing model forward passes while maintaining the visual quality of generated images.

PDF Details

ICLR Conference 2025 Conference Paper

Your Absorbing Discrete Diffusion Secretly Models the Conditional Distributions of Clean Data

  • Jingyang Ou
  • Shen Nie
  • Kaiwen Xue
  • Fengqi Zhu
  • Jiacheng Sun
  • Zhenguo Li
  • Chongxuan Li

Discrete diffusion models with absorbing processes have shown promise in language modeling. The key quantities to be estimated are the ratios between the marginal probabilities of two transitive states at all timesteps, called the concrete score. In this paper, we reveal that the concrete score in absorbing diffusion can be expressed as conditional probabilities of clean data, multiplied by a time-dependent scalar in an analytic form. Motivated by this finding, we propose reparameterized absorbing discrete diffusion (RADD), a dedicated diffusion model without time-condition that characterizes the time-independent conditional probabilities. Besides its simplicity, RADD can reduce the number of function evaluations (NFEs) by caching the output of the time-independent network when the noisy sample remains unchanged in a sampling interval, which enables sampling acceleration. Built upon the new perspective of conditional distributions, we further unify absorbing discrete diffusion and any-order autoregressive models (AO-ARMs), showing that the upper bound on the negative log-likelihood for the diffusion model can be interpreted as an expected negative log-likelihood for AO-ARMs. Further, our RADD models achieve SOTA performance among diffusion models on 5 zero-shot language modeling benchmarks (measured by perplexity) at the GPT-2 scale. Our code is available at \url{https://github.com/ML-GSAI/RADD}.

Details

NeurIPS Conference 2024 Conference Paper

DAPE: Data-Adaptive Positional Encoding for Length Extrapolation

  • Chuanyang Zheng
  • Yihang Gao
  • Han Shi
  • Minbin Huang
  • Jingyao Li
  • Jing Xiong
  • Xiaozhe Ren
  • Michael Ng

Positional encoding plays a crucial role in transformers, significantly impact- ing model performance and length generalization. Prior research has introduced absolute positional encoding (APE) and relative positional encoding (RPE) to distinguish token positions in given sequences. However, both APE and RPE remain fixed after model training regardless of input data, limiting their adaptability and flexibility. Hence, we expect that the desired positional encoding should be data-adaptive and can be dynamically adjusted with the given attention. In this paper, we propose a Data-Adaptive Positional Encoding (DAPE) method, which dynamically and semantically adjusts based on input context and learned fixed priors. Experimental validation on real-world datasets (Arxiv, Books3, and CHE) demonstrates that DAPE enhances model performances in terms of trained length and length generalization, where the improvements are statistically significant. The model visualization suggests that our model can keep both local and anti-local information. Finally, we successfully train the model on sequence length 128 and achieve better performance at evaluation sequence length 8192, compared with other static positional encoding methods, revealing the benefit of the adaptive positional encoding method.

PDF Details DOI

AAAI Conference 2024 Conference Paper

DeepAccident: A Motion and Accident Prediction Benchmark for V2X Autonomous Driving

  • Tianqi Wang
  • Sukmin Kim
  • Ji Wenxuan
  • Enze Xie
  • Chongjian Ge
  • Junsong Chen
  • Zhenguo Li
  • Ping Luo

Safety is the primary priority of autonomous driving. Nevertheless, no published dataset currently supports the direct and explainable safety evaluation for autonomous driving. In this work, we propose DeepAccident, a large-scale dataset generated via a realistic simulator containing diverse accident scenarios that frequently occur in real-world driving. The proposed DeepAccident dataset includes 57K annotated frames and 285K annotated samples, approximately 7 times more than the large-scale nuScenes dataset with 40k annotated samples. In addition, we propose a new task, end-to-end motion and accident prediction, which can be used to directly evaluate the accident prediction ability for different autonomous driving algorithms. Furthermore, for each scenario, we set four vehicles along with one infrastructure to record data, thus providing diverse viewpoints for accident scenarios and enabling V2X (vehicle-to-everything) research on perception and prediction tasks. Finally, we present a baseline V2X model named V2XFormer that demonstrates superior performance for motion and accident prediction and 3D object detection compared to the single-vehicle model.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

Diffusion of Thought: Chain-of-Thought Reasoning in Diffusion Language Models

  • Jiacheng Ye
  • Shansan Gong
  • Liheng Chen
  • Lin Zheng
  • Jiahui Gao
  • Han Shi
  • Chuan Wu
  • Xin Jiang

Recently, diffusion models have garnered significant interest in the field of text processing due to their many potential advantages compared to conventional autoregressive models. In this work, we propose Diffusion-of-Thought (DoT), a novel approach that integrates diffusion models with Chain-of-Thought, a well-established technique for improving the reasoning ability of autoregressive language models. In contrast to autoregressive language models that make decisions in a left-to-right, token-by-token manner, DoT allows reasoning steps to diffuse over time through a diffusion language model and offers greater flexibility in trading-off computation for reasoning performance. Our experimental results demonstrate the effectiveness of DoT in multi-digit multiplication, boolean logic, and grade school math problems. In addition to that, DoT showcases promising self-correction abilities and benefits from existing reasoning-enhancing techniques like self-consistency decoding. Our findings contribute to the understanding and development of reasoning with diffusion language models.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

Dual Risk Minimization: Towards Next-Level Robustness in Fine-tuning Zero-Shot Models

  • Kaican Li
  • Weiyan Xie
  • Yongxiang Huang
  • Didan Deng
  • Lanqing Hong
  • Zhenguo Li
  • Ricardo Silva
  • Nevin L. Zhang

Fine-tuning foundation models often compromises their robustness to distribution shifts. To remedy this, most robust fine-tuning methods aim to preserve the pre-trained features. However, not all pre-trained features are robust and those methods are largely indifferent to which ones to preserve. We propose dual risk minimization (DRM), which combines empirical risk minimization with worst-case risk minimization, to better preserve the core features of downstream tasks. In particular, we utilize core-feature descriptions generated by LLMs to induce core-based zero-shot predictions which then serve as proxies to estimate the worst-case risk. DRM balances two crucial aspects of model robustness: expected performance and worst-case performance, establishing a new state of the art on various real-world benchmarks. DRM significantly improves the out-of-distribution performance of CLIP ViT-L/14@336 on ImageNet (75. 9$\to$77. 1), WILDS-iWildCam (47. 1$\to$51. 8), and WILDS-FMoW (50. 7$\to$53. 1); opening up new avenues for robust fine-tuning. Our code is available at https: //github. com/vaynexie/DRM.

PDF Details DOI

ICLR Conference 2024 Conference Paper

Gaining Wisdom from Setbacks: Aligning Large Language Models via Mistake Analysis

  • Kai Chen 0023
  • Chunwei Wang
  • Kuo Yang
  • Jianhua Han
  • Lanqing Hong
  • Fei Mi
  • Hang Xu 0004
  • Zhengying Liu

The rapid development of large language models (LLMs) has not only provided numerous opportunities but also presented significant challenges. This becomes particularly evident when LLMs inadvertently generate harmful or toxic content, either unintentionally or because of intentional inducement. Existing alignment methods usually direct LLMs toward the favorable outcomes by utilizing human-annotated, flawless instruction-response pairs. Conversely, this study proposes a novel alignment technique based on mistake analysis, which deliberately exposes LLMs to erroneous content to learn the reasons for mistakes and how to avoid them. In this case, mistakes are repurposed into valuable data for alignment, effectively helping to avoid the production of erroneous responses. Without external models or human annotations, our method leverages a model's intrinsic ability to discern undesirable mistakes and improves the safety of its generated responses. Experimental results reveal that our method outperforms existing alignment approaches in enhancing model safety while maintaining the overall utility.

Details

NeurIPS Conference 2024 Conference Paper

GenArtist: Multimodal LLM as an Agent for Unified Image Generation and Editing

  • Zhenyu Wang
  • Aoxue Li
  • Zhenguo Li
  • Xihui Liu

Despite the success achieved by existing image generation and editing methods, current models still struggle with complex problems including intricate text prompts, and the absence of verification and self-correction mechanisms makes the generated images unreliable. Meanwhile, a single model tends to specialize in particular tasks and possess the corresponding capabilities, making it inadequate for fulfilling all user requirements. We propose GenArtist, a unified image generation and editing system, coordinated by a multimodal large language model (MLLM) agent. We integrate a comprehensive range of existing models into the tool library and utilize the agent for tool selection and execution. For a complex problem, the MLLM agent decomposes it into simpler sub-problems and constructs a tree structure to systematically plan the procedure of generation, editing, and self-correction with step-by-step verification. By automatically generating missing position-related inputs and incorporating position information, the appropriate tool can be effectively employed to address each sub-problem. Experiments demonstrate that GenArtist can perform various generation and editing tasks, achieving state-of-the-art performance and surpassing existing models such as SDXL and DALL-E 3, as can be seen in Fig. 1. We will open-source the code for future research and applications.

PDF Details DOI

ICLR Conference 2024 Conference Paper

GeoDiffusion: Text-Prompted Geometric Control for Object Detection Data Generation

  • Kai Chen 0023
  • Enze Xie
  • Zhe Chen
  • Yibo Wang 0039
  • Lanqing Hong
  • Zhenguo Li
  • Dit-Yan Yeung

Diffusion models have attracted significant attention due to the remarkable ability to create content and generate data for tasks like image classification. However, the usage of diffusion models to generate the high-quality object detection data remains an underexplored area, where not only image-level perceptual quality but also geometric conditions such as bounding boxes and camera views are essential. Previous studies have utilized either copy-paste synthesis or layout-to-image (L2I) generation with specifically designed modules to encode the semantic layouts. In this paper, we propose the GeoDiffusion, a simple framework that can flexibly translate various geometric conditions into text prompts and empower pre-trained text-to-image (T2I) diffusion models for high-quality detection data generation. Unlike previous L2I methods, our GeoDiffusion is able to encode not only the bounding boxes but also extra geometric conditions such as camera views in self-driving scenes. Extensive experiments demonstrate GeoDiffusion outperforms previous L2I methods while maintaining 4x training time faster. To the best of our knowledge, this is the first work to adopt diffusion models for layout-to-image generation with geometric conditions and demonstrate that L2I-generated images can be beneficial for improving the performance of object detectors.

Details

ICLR Conference 2024 Conference Paper

Large Language Models as Automated Aligners for benchmarking Vision-Language Models

  • Yuanfeng Ji
  • Chongjian Ge
  • Weikai Kong
  • Enze Xie
  • Zhengying Liu
  • Zhenguo Li
  • Ping Luo 0002

With the advancements in Large Language Models (LLMs), Vision-Language Models (VLMs) have reached a new level of sophistication, showing notable competence in executing intricate cognition and reasoning tasks. However, existing evaluation benchmarks, primarily relying on rigid, hand-crafted datasets to measure task-specific performance, face significant limitations in assessing the alignment of these increasingly anthropomorphic models with human intelligence. In this work, we address the limitations via Auto-Bench, which delves into exploring LLMs as proficient aligners, measuring the alignment between VLMs and human intelligence and value through automatic data curation and assessment. Specifically, for data curation, Auto-Bench utilizes LLMs (e.g., GPT-4) to automatically generate a vast set of question-answer-reasoning triplets via prompting on visual symbolic representations (e.g., captions, object locations, instance relationships, and etc. The curated data closely matches human intent, owing to the extensive world knowledge embedded in LLMs. Through this pipeline, a total of 28.5K human-verified and 3,504K unfiltered question-answer-reasoning triplets have been curated, covering 4 primary abilities and 16 sub-abilities. We subsequently engage LLMs like GPT-3.5 to serve as judges, implementing the quantitative and qualitative automated assessments to facilitate a comprehensive evaluation of VLMs. Our validation results reveal that LLMs are proficient in both evaluation data curation and model assessment, achieving an average agreement rate of 85%. We envision Auto-Bench as a flexible, scalable, and comprehensive benchmark for evaluating the evolving sophisticated VLMs.

Details

ICLR Conference 2024 Conference Paper

LEGO-Prover: Neural Theorem Proving with Growing Libraries

  • Haiming Wang
  • Huajian Xin
  • Chuanyang Zheng
  • Zhengying Liu
  • Qingxing Cao
  • Yinya Huang
  • Jing Xiong
  • Han Shi

Despite the success of large language models (LLMs), the task of theorem proving still remains one of the hardest reasoning tasks that is far from being fully solved. Prior methods using language models have demonstrated promising results, but they still struggle to prove even middle school level theorems. One common limitation of these methods is that they assume a fixed theorem library during the whole theorem proving process. However, as we all know, creating new useful theorems or even new theories is not only helpful but crucial and necessary for advancing mathematics and proving harder and deeper results. In this work, we present LEGO-Prover, which employs a growing skill library containing verified lemmas as skills to augment the capability of LLMs used in theorem proving. By constructing the proof modularly, LEGO-Prover enables LLMs to utilize existing skills retrieved from the library and to create new skills during the proving process. These skills are further evolved (by prompting an LLM) to enrich the library on another scale. Modular and reusable skills are constantly added to the library to enable tackling increasingly intricate mathematical problems. Moreover, the learned library further bridges the gap between human proofs and formal proofs by making it easier to impute missing steps. LEGO-Prover advances the state-of-the-art pass rate on miniF2F-valid (48.0\% to 57.0\%) and miniF2F-test (45.5\% to 50.0\%). During the proving process, LEGO-Prover also generates over 20,000 skills (theorems/lemmas) and adds them to the growing library. Our ablation study indicates that these newly added skills are indeed helpful for proving theorems, resulting in a 4.9\% improvement in success rate

Details

TMLR Journal 2024 Journal Article

Lyra: Orchestrating Dual Correction in Automated Theorem Proving

  • Chuanyang Zheng
  • Haiming Wang
  • Enze Xie
  • Zhengying Liu
  • Jiankai Sun
  • Huajian Xin
  • Jianhao Shen
  • Zhenguo Li

Large Language Models (LLMs) present an intriguing avenue for exploration in the field of formal theorem proving. Nevertheless, their full potential, particularly concerning the mitigation of hallucinations and refinement through prover error messages, remains an area that has yet to be thoroughly investigated. To enhance the effectiveness of LLMs in the field, we introduce the Lyra, a new framework that employs two distinct correction mechanisms: Tool Correction (TC) and Conjecture Correction (CC). To implement Tool Correction in the post-processing of formal proofs, we leverage prior knowledge to utilize predefined prover tools (e.g., Sledgehammer) for guiding the replacement of incorrect tools. Tool Correction significantly contributes to mitigating hallucinations, thereby improving the overall accuracy of the proof. In addition, we introduce Conjecture Correction, an error feedback mechanism designed to interact with prover to refine formal proof conjectures with prover error messages. Compared to the previous refinement framework, the proposed Conjecture Correction refines generation with instruction but does not collect paired (generation, error & refinement) prompts. Our method has achieved state-of-the-art (SOTA) performance on both miniF2F validation (48.0% → 55.3%) and test (45.5% → 51.2%). We also present 3 IMO problems solved by Lyra. We believe Tool Correction (post-process for hallucination mitigation) and Conjecture Correction (subgoal adjustment from interaction with the environment) could provide a promising avenue for future research in this field.

PDF Details

ICLR Conference 2024 Conference Paper

MagicDrive: Street View Generation with Diverse 3D Geometry Control

  • Ruiyuan Gao 0001
  • Kai Chen 0023
  • Enze Xie
  • Lanqing Hong
  • Zhenguo Li
  • Dit-Yan Yeung
  • Qiang Xu 0001

Recent advancements in diffusion models have significantly enhanced the data synthesis with 2D control. Yet, precise 3D control in street view generation, crucial for 3D perception tasks, remains elusive. Specifically, utilizing Bird's-Eye View (BEV) as the primary condition often leads to challenges in geometry control (e.g., height), affecting the representation of object shapes, occlusion patterns, and road surface elevations, all of which are essential to perception data synthesis, especially for 3D object detection tasks. In this paper, we introduce MagicDrive, a novel street view generation framework, offering diverse 3D geometry controls including camera poses, road maps, and 3D bounding boxes, together with textual descriptions, achieved through tailored encoding strategies. Besides, our design incorporates a cross-view attention module, ensuring consistency across multiple camera views. With MagicDrive, we achieve high-fidelity street-view image & video synthesis that captures nuanced 3D geometry and various scene descriptions, enhancing tasks like BEV segmentation and 3D object detection. Project Website: https://flymin.github.io/magicdrive

Details

ICLR Conference 2024 Conference Paper

MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models

  • Longhui Yu
  • Weisen Jiang
  • Han Shi
  • Jincheng Yu
  • Zhengying Liu
  • Yu Zhang 0006
  • James T. Kwok
  • Zhenguo Li

Large language models (LLMs) have pushed the limits of natural language understanding and exhibited excellent problem-solving ability. Despite the great success, most existing open-source LLMs (\eg, LLaMA-2) are still far away from satisfactory for solving mathematical problems due to the complex reasoning procedures. To bridge this gap, we propose \emph{MetaMath}, a finetuned language model that specializes in mathematical reasoning. Specifically, we start by bootstrapping mathematical questions by rewriting the question from multiple perspectives, which results in a new dataset called MetaMathQA. Then we finetune the LLaMA-2 models on MetaMathQA. Experimental results on two popular benchmarks (\ie, GSM8K and MATH) for mathematical reasoning demonstrate that MetaMath outperforms a suite of open-source LLMs by a significant margin. Our MetaMath-7B model achieves $66.5\%$ on GSM8K and $19.8\%$ on MATH, exceeding the state-of-the-art models of the same size by $11.5\%$ and $8.7\%$. Particularly, MetaMath-70B achieves an accuracy of $82.3\%$ on GSM8K, slightly better than GPT-3.5-Turbo. We release the MetaMathQA dataset, the MetaMath models with different model sizes and the training code for public use.

Details

ICLR Conference 2024 Conference Paper

MUSTARD: Mastering Uniform Synthesis of Theorem and Proof Data

  • Yinya Huang
  • Xiaohan Lin
  • Zhengying Liu
  • Qingxing Cao
  • Huajian Xin
  • Haiming Wang
  • Zhenguo Li
  • Linqi Song

Recent large language models (LLMs) have witnessed significant advancement in various tasks, including mathematical reasoning and theorem proving. As these two tasks require strict and formal multi-step inference, they are appealing domains for exploring the reasoning ability of LLMs but still face important challenges. Previous studies such as Chain-of-Thought (CoT) have revealed the effectiveness of intermediate steps guidance. However, such step-wise annotation requires heavy labor, leading to insufficient training steps for current benchmarks. To fill this gap, this work introduces MUSTARD, a data generation framework that masters uniform synthesis of theorem and proof data of high quality and diversity. MUSTARD synthesizes data in three stages: (1) It samples a few mathematical concept seeds as the problem category. (2) Then, it prompts a generative language model with the sampled concepts to obtain both the problems and their step-wise formal solutions. (3) Lastly, the framework utilizes a proof assistant (e.g., Lean Prover) to filter the valid proofs. With the proposed MUSTARD, we present a theorem-and-proof benchmark MUSTARDSAUCE with 5,866 valid data points. Each data point contains an informal statement, an informal proof, and a translated formal proof that passes the prover validation. We perform extensive analysis and demonstrate that MUSTARD generates validated high-quality step-by-step data. We further apply the MUSTARDSAUCE for fine-tuning smaller language models. The fine-tuned Llama 2-7B achieves a 15.41% average relative performance gain in automated theorem proving, and 8.18% in math word problems. Codes and data are available at https://github.com/Eleanor-H/MUSTARD.

Details

ICLR Conference 2024 Conference Paper

PixArt-α: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis

  • Junsong Chen
  • Jincheng Yu
  • Chongjian Ge
  • Lewei Yao
  • Enze Xie
  • Zhongdao Wang
  • James T. Kwok
  • Ping Luo 0002

The most advanced text-to-image (T2I) models require significant training costs (e.g., millions of GPU hours), seriously hindering the fundamental innovation for the AIGC community while increasing CO2 emissions. This paper introduces PixArt-$\alpha$, a Transformer-based T2I diffusion model whose image generation quality is competitive with state-of-the-art image generators (e.g., Imagen, SDXL, and even Midjourney), reaching near-commercial application standards. Additionally, it supports high-resolution image synthesis up to 1024px resolution with low training cost, as shown in Figure 1 and 2. To achieve this goal, three core designs are proposed: (1) Training strategy decomposition: We devise three distinct training steps that separately optimize pixel dependency, text-image alignment, and image aesthetic quality; (2) Efficient T2I Transformer: We incorporate cross-attention modules into Diffusion Transformer (DiT) to inject text conditions and streamline the computation-intensive class-condition branch; (3) High-informative data: We emphasize the significance of concept density in text-image pairs and leverage a large Vision-Language model to auto-label dense pseudo-captions to assist text-image alignment learning. As a result, PixArt-$\alpha$'s training speed markedly surpasses existing large-scale T2I models, e.g., PixArt-$\alpha$ only takes 10.8% of Stable Diffusion v1.5's training time (~675 vs. ~6,250 A100 GPU days), saving nearly \\$300,000 (\\$26,000 vs. \\$320,000) and reducing 90% CO2 emissions. Moreover, compared with a larger SOTA model, RAPHAEL, our training cost is merely 1%. Extensive experiments demonstrate that PixArt-$\alpha$ excels in image quality, artistry, and semantic control. We hope PixArt-$\alpha$ will provide new insights to the AIGC community and startups to accelerate building their own high-quality yet low-cost generative models from scratch.

Details

NeurIPS Conference 2024 Conference Paper

Proving Theorems Recursively

  • Haiming Wang
  • Huajian Xin
  • Zhengying Liu
  • Wenda Li
  • Yinya Huang
  • Jianqiao Lu
  • Zhicheng Yang
  • Jing Tang

Recent advances in automated theorem proving leverages language models to explore expanded search spaces by step-by-step proof generation. However, such approaches are usually based on short-sighted heuristics (e. g. , log probability or value function scores) that potentially lead to suboptimal or even distracting subgoals, preventing us from finding longer proofs. To address this challenge, we propose POETRY (PrOvE Theorems RecursivelY), which proves theorems in a recursive, level-by-level manner in the Isabelle theorem prover. Unlike previous step-by-step methods, POETRY searches for a verifiable sketch of the proof at each level and focuses on solving the current level's theorem or conjecture. Detailed proofs of intermediate conjectures within the sketch are temporarily replaced by a placeholder tactic called sorry, deferring their proofs to subsequent levels. This approach allows the theorem to be tackled incrementally by outlining the overall theorem at the first level and then solving the intermediate conjectures at deeper levels. Experiments are conducted on the miniF2F and PISA datasets and significant performance gains are observed in our POETRY approach over state-of-the-art methods. POETRY on miniF2F achieves an average proving success rate improvement of 5. 1%. Moreover, we observe a substantial increase in the maximum proof length found by POETRY, from 10 to 26.

PDF Details DOI

ICML Conference 2024 Conference Paper

The Surprising Effectiveness of Skip-Tuning in Diffusion Sampling

  • Jiajun Ma
  • Shuchen Xue
  • Tianyang Hu 0001
  • Wenjia Wang
  • Zhaoqiang Liu
  • Zhenguo Li
  • Zhiming Ma
  • Kenji Kawaguchi

With the incorporation of the UNet architecture, diffusion probabilistic models have become a dominant force in image generation tasks. One key design in UNet is the skip connections between the encoder and decoder blocks. Although skip connections have been shown to improve training stability and model performance, we point out that such shortcuts can be a limiting factor for the complexity of the transformation. As the sampling steps decrease, the generation process and the role of the UNet get closer to the push-forward transformations from Gaussian distribution to the target, posing a challenge for the network’s complexity. To address this challenge, we propose Skip-Tuning, a simple yet surprisingly effective training-free tuning method on the skip connections. For instance, our method can achieve 100% FID improvement for pretrained EDM on ImageNet 64 with only 19 NFEs (1. 75), breaking the limit of ODE samplers regardless of sampling steps. Surprisingly, the improvement persists when we increase the number of sampling steps and can even surpass the best result from EDM-2 (1. 58) with only 39 NFEs (1. 57). Comprehensive exploratory experiments are conducted to shed light on the surprising effectiveness of our Skip-Tuning. We observe that while Skip-Tuning increases the score-matching losses in the pixel space, the losses in the feature space are reduced, particularly at intermediate noise levels, which coincide with the most effective range accounting for image quality improvement.

Details

NeurIPS Conference 2024 Conference Paper

Towards Understanding the Working Mechanism of Text-to-Image Diffusion Model

  • Mingyang Yi
  • Aoxue Li
  • Yi Xin
  • Zhenguo Li

Recently, the strong latent Diffusion Probabilistic Model (DPM) has been applied to high-quality Text-to-Image (T2I) generation (e. g. , Stable Diffusion), by injecting the encoded target text prompt into the gradually denoised diffusion image generator. Despite the success of DPM in practice, the mechanism behind it remains to be explored. To fill this blank, we begin by examining the intermediate statuses during the gradual denoising generation process in DPM. The empirical observations indicate, the shape of image is reconstructed after the first few denoising steps, and then the image is filled with details (e. g. , texture). The phenomenon is because the low-frequency signal (shape relevant) of the noisy image is not corrupted until the final stage in the forward process (initial stage of generation) of adding noise in DPM. Inspired by the observations, we proceed to explore the influence of each token in the text prompt during the two stages. After a series of experiments of T2I generations conditioned on a set of text prompts. We conclude that in the earlier generation stage, the image is mostly decided by the special token [\texttt{EOS}] in the text prompt, and the information in the text prompt is already conveyed in this stage. After that, the diffusion model completes the details of generated images by information from themselves. Finally, we propose to apply this observation to accelerate the process of T2I generation by properly removing text guidance, which finally accelerates the sampling up to 25\%+.

PDF Details DOI

ICLR Conference 2023 Conference Paper

Breaking Correlation Shift via Conditional Invariant Regularizer

  • Mingyang Yi
  • Ruoyu Wang 0016
  • Jiacheng Sun
  • Zhenguo Li
  • Zhiming Ma

Recently, generalization on out-of-distribution (OOD) data with correlation shift has attracted great attentions. The correlation shift is caused by the spurious attributes that correlate to the class label, as the correlation between them may vary in training and test data. For such a problem, we show that given the class label, the models that are conditionally independent of spurious attributes are OOD generalizable. Based on this, a metric Conditional Spurious Variation (CSV) which controls the OOD generalization error, is proposed to measure such conditional independence. To improve the OOD generalization, we regularize the training process with the proposed CSV. Under mild assumptions, our training objective can be formulated as a nonconvex-concave mini-max problem. An algorithm with a provable convergence rate is proposed to solve the problem. Extensive empirical results verify our algorithm's efficacy in improving OOD generalization.

Details

ICLR Conference 2023 Conference Paper

CO3: Cooperative Unsupervised 3D Representation Learning for Autonomous Driving

  • Runjian Chen
  • Yao Mu 0001
  • Runsen Xu
  • Wenqi Shao
  • Chenhan Jiang
  • Hang Xu 0004
  • Yu Qiao 0001
  • Zhenguo Li

Unsupervised contrastive learning for indoor-scene point clouds has achieved great successes. However, unsupervised representation learning on outdoor-scene point clouds remains challenging because previous methods need to reconstruct the whole scene and capture partial views for the contrastive objective. This is infeasible in outdoor scenes with moving objects, obstacles, and sensors. In this paper, we propose CO3, namely {Co}operative {Co}ntrastive Learning and {Co}ntextual Shape Prediction, to learn 3D representation for outdoor-scene point clouds in an unsupervised manner. CO3 has several merits compared to existing methods. (1) It utilizes LiDAR point clouds from vehicle-side and infrastructure-side to build views that differ enough but meanwhile maintain common semantic information for contrastive learning, which are more appropriate than views built by previous methods. (2) Alongside the contrastive objective, we propose contextual shape prediction to bring more task-relevant information for unsupervised 3D point cloud representation learning and we also provide a theoretical analysis for this pre-training goal. (3) As compared to previous methods, representation learned by CO3 is able to be transferred to different outdoor scene dataset collected by different type of LiDAR sensors. (4) CO3 improves current state-of-the-art methods on Once, KITTI and NuScenes datasets by up to 2.58 mAP in 3D object detection task and 3.54 mIoU in LiDAR semantic segmentation task. Codes and models will be released.

Details

NeurIPS Conference 2023 Conference Paper

Complexity Matters: Rethinking the Latent Space for Generative Modeling

  • Tianyang Hu
  • Fei Chen
  • Haonan Wang
  • Jiawei Li
  • Wenjia Wang
  • Jiacheng Sun
  • Zhenguo Li

In generative modeling, numerous successful approaches leverage a low-dimensional latent space, e. g. , Stable Diffusion models the latent space induced by an encoder and generates images through a paired decoder. Although the selection of the latent space is empirically pivotal, determining the optimal choice and the process of identifying it remain unclear. In this study, we aim to shed light on this under-explored topic by rethinking the latent space from the perspective of model complexity. Our investigation starts with the classic generative adversarial networks (GANs). Inspired by the GAN training objective, we propose a novel "distance" between the latent and data distributions, whose minimization coincides with that of the generator complexity. The minimizer of this distance is characterized as the optimal data-dependent latent that most effectively capitalizes on the generator's capacity. Then, we consider parameterizing such a latent distribution by an encoder network and propose a two-stage training strategy called Decoupled Autoencoder (DAE), where the encoder is only updated in the first stage with an auxiliary decoder and then frozen in the second stage while the actual decoder is being trained. DAE can improve the latent distribution and as a result, improve the generative performance. Our theoretical analyses are corroborated by comprehensive experiments on various models such as VQGAN and Diffusion Transformer, where our modifications yield significant improvements in sample quality with decreased model complexity.

PDF Details

AAAI Conference 2023 Conference Paper

DAMix: Exploiting Deep Autoregressive Model Zoo for Improving Lossless Compression Generalization

  • Qishi Dong
  • Fengwei Zhou
  • Ning Kang
  • Chuanlong Xie
  • Shifeng Zhang
  • Jiawei Li
  • Heng Peng
  • Zhenguo Li

Deep generative models have demonstrated superior performance in lossless compression on identically distributed data. However, in real-world scenarios, data to be compressed are of various distributions and usually cannot be known in advance. Thus, commercially expected neural compression must have strong Out-of-Distribution (OoD) generalization capabilities. Compared with traditional compression methods, deep learning methods have intrinsic flaws for OoD generalization. In this work, we make the attempt to tackle this challenge via exploiting a zoo of Deep Autoregressive models (DAMix). We build a model zoo consisting of autoregressive models trained on data from diverse distributions. In the test phase, we select useful expert models by a simple model evaluation score and adaptively aggregate the predictions of selected models. By assuming the outputs from each expert model are biased in favor of their training distributions, a von Mises-Fisher based filter is proposed to recover the value of unbiased predictions that provides more accurate density estimations than a single model. We derive the posterior of unbiased predictions as well as concentration parameters in the filter, and a novel temporal Stein variational gradient descent for sequential data is proposed to adaptively update the posterior distributions. We evaluate DAMix on 22 image datasets, including in-distribution and OoD data, and demonstrate that making use of unbiased predictions has up to 45.6% improvement over the single model trained on ImageNet.

PDF Details DOI

NeurIPS Conference 2023 Conference Paper

Diff-Instruct: A Universal Approach for Transferring Knowledge From Pre-trained Diffusion Models

  • Weijian Luo
  • Tianyang Hu
  • Shifeng Zhang
  • Jiacheng Sun
  • Zhenguo Li
  • Zhihua Zhang

Due to the ease of training, ability to scale, and high sample quality, diffusion models (DMs) have become the preferred option for generative modeling, with numerous pre-trained models available for a wide variety of datasets. Containing intricate information about data distributions, pre-trained DMs are valuable assets for downstream applications. In this work, we consider learning from pre-trained DMs and transferring their knowledge to other generative models in a data-free fashion. Specifically, we propose a general framework called Diff-Instruct to instruct the training of arbitrary generative models as long as the generated samples are differentiable with respect to the model parameters. Our proposed Diff-Instruct is built on a rigorous mathematical foundation where the instruction process directly corresponds to minimizing a novel divergence we call Integral Kullback-Leibler (IKL) divergence. IKL is tailored for DMs by calculating the integral of the KL divergence along a diffusion process, which we show to be more robust in comparing distributions with misaligned supports. We also reveal non-trivial connections of our method to existing works such as DreamFusion \citep{poole2022dreamfusion}, and generative adversarial training. To demonstrate the effectiveness and universality of Diff-Instruct, we consider two scenarios: distilling pre-trained diffusion models and refining existing GAN models. The experiments on distilling pre-trained diffusion models show that Diff-Instruct results in state-of-the-art single-step diffusion-based models. The experiments on refining GAN models show that the Diff-Instruct can consistently improve the pre-trained generators of GAN models across various settings. Our official code is released through \url{https: //github. com/pkulwj1994/diff_instruct}.

PDF Details

NeurIPS Conference 2023 Conference Paper

DiffComplete: Diffusion-based Generative 3D Shape Completion

  • Ruihang Chu
  • Enze Xie
  • Shentong Mo
  • Zhenguo Li
  • Matthias Niessner
  • Chi-Wing Fu
  • Jiaya Jia

We introduce a new diffusion-based approach for shape completion on 3D range scans. Compared with prior deterministic and probabilistic methods, we strike a balance between realism, multi-modality, and high fidelity. We propose DiffComplete by casting shape completion as a generative task conditioned on the incomplete shape. Our key designs are two-fold. First, we devise a hierarchical feature aggregation mechanism to inject conditional features in a spatially-consistent manner. So, we can capture both local details and broader contexts of the conditional inputs to control the shape completion. Second, we propose an occupancy-aware fusion strategy in our model to enable the completion of multiple partial shapes and introduce higher flexibility on the input conditions. DiffComplete sets a new SOTA performance (e. g. , 40% decrease on $l_1$ error) on two large-scale 3D shape completion benchmarks. Our completed shapes not only have a realistic outlook compared with the deterministic methods but also exhibit high similarity to the ground truths compared with the probabilistic alternatives. Further, DiffComplete has strong generalizability on objects of entirely unseen classes for both synthetic and real data, eliminating the need for model re-training in various applications.

PDF Details

NeurIPS Conference 2023 Conference Paper

DiT-3D: Exploring Plain Diffusion Transformers for 3D Shape Generation

  • Shentong Mo
  • Enze Xie
  • Ruihang Chu
  • Lanqing Hong
  • Matthias Niessner
  • Zhenguo Li

Recent Diffusion Transformers (i. e. , DiT) have demonstrated their powerful effectiveness in generating high-quality 2D images. However, it is unclear how the Transformer architecture performs equally well in 3D shape generation, as previous 3D diffusion methods mostly adopted the U-Net architecture. To bridge this gap, we propose a novel Diffusion Transformer for 3D shape generation, named DiT-3D, which can directly operate the denoising process on voxelized point clouds using plain Transformers. Compared to existing U-Net approaches, our DiT-3D is more scalable in model size and produces much higher quality generations. Specifically, the DiT-3D adopts the design philosophy of DiT but modifies it by incorporating 3D positional and patch embeddings to aggregate input from voxelized point clouds. To reduce the computational cost of self-attention in 3D shape generation, we incorporate 3D window attention into Transformer blocks, as the increased 3D token length resulting from the additional dimension of voxels can lead to high computation. Finally, linear and devoxelization layers are used to predict the denoised point clouds. In addition, we empirically observe that the pre-trained DiT-2D checkpoint on ImageNet can significantly improve DiT-3D on ShapeNet. Experimental results on the ShapeNet dataset demonstrate that the proposed DiT-3D achieves state-of-the-art performance in high-fidelity and diverse 3D point cloud generation.

PDF Details

ICML Conference 2023 Conference Paper

Explore and Exploit the Diverse Knowledge in Model Zoo for Domain Generalization

  • Yimeng Chen
  • Tianyang Hu 0001
  • Fengwei Zhou
  • Zhenguo Li
  • Zhiming Ma

The proliferation of pretrained models, as a result of advancements in pretraining techniques, has led to the emergence of a vast zoo of publicly available models. Effectively utilizing these resources to obtain models with robust out-of-distribution generalization capabilities for downstream tasks has become a crucial area of research. Previous research has primarily focused on identifying the most powerful models within the model zoo, neglecting to fully leverage the diverse inductive biases contained within. This paper argues that the knowledge contained in weaker models is valuable and presents a method for leveraging the diversity within the model zoo to improve out-of-distribution generalization capabilities. Specifically, we investigate the behaviors of various pretrained models across different domains of downstream tasks by characterizing the variations in their encoded representations in terms of two dimensions: diversity shift and correlation shift. This characterization enables us to propose a new algorithm for integrating diverse pretrained models, not limited to the strongest models, in order to achieve enhanced out-of-distribution generalization performance. Our proposed method demonstrates state-of-the-art empirical results on a variety of datasets, thus validating the benefits of utilizing diverse knowledge.

Details

AAAI Conference 2023 Conference Paper

Fair-CDA: Continuous and Directional Augmentation for Group Fairness

  • Rui Sun
  • Fengwei Zhou
  • Zhenhua Dong
  • Chuanlong Xie
  • Lanqing Hong
  • Jiawei Li
  • Rui Zhang
  • Zhen Li

In this work, we propose Fair-CDA, a fine-grained data augmentation strategy for imposing fairness constraints. We use a feature disentanglement method to extract the features highly related to the sensitive attributes. Then we show that group fairness can be achieved by regularizing the models on transition paths of sensitive features between groups. By adjusting the perturbation strength in the direction of the paths, our proposed augmentation is controllable and auditable. To alleviate the accuracy degradation caused by fairness constraints, we further introduce a calibrated model to impute labels for the augmented data. Our proposed method does not assume any data generative model and ensures good generalization for both accuracy and fairness. Experimental results show that Fair-CDA consistently outperforms state-of-the-art methods on widely-used benchmarks, e.g., Adult, CelebA and MovieLens. Especially, Fair-CDA obtains an 86.3% relative improvement for fairness while maintaining the accuracy on the Adult dataset. Moreover, we evaluate Fair-CDA in an online recommendation system to demonstrate the effectiveness of our method in terms of accuracy and fairness.

PDF Details DOI

NeurIPS Conference 2023 Conference Paper

SA-Solver: Stochastic Adams Solver for Fast Sampling of Diffusion Models

  • Shuchen Xue
  • Mingyang Yi
  • Weijian Luo
  • Shifeng Zhang
  • Jiacheng Sun
  • Zhenguo Li
  • Zhi-Ming Ma

Diffusion Probabilistic Models (DPMs) have achieved considerable success in generation tasks. As sampling from DPMs is equivalent to solving diffusion SDE or ODE which is time-consuming, numerous fast sampling methods built upon improved differential equation solvers are proposed. The majority of such techniques consider solving the diffusion ODE due to its superior efficiency. However, stochastic sampling could offer additional advantages in generating diverse and high-quality data. In this work, we engage in a comprehensive analysis of stochastic sampling from two aspects: variance-controlled diffusion SDE and linear multi-step SDE solver. Based on our analysis, we propose SA-Solver, which is an improved efficient stochastic Adams method for solving diffusion SDE to generate data with high quality. Our experiments show that SA-Solver achieves: 1) improved or comparable performance compared with the existing state-of-the-art (SOTA) sampling methods for few-step sampling; 2) SOTA FID on substantial benchmark datasets under a suitable number of function evaluations (NFEs).

PDF Details

ICLR Conference 2023 Conference Paper

Self-Guided Noise-Free Data Generation for Efficient Zero-Shot Learning

  • Jiahui Gao
  • Renjie Pi
  • Yong Lin
  • Hang Xu 0004
  • Jiacheng Ye
  • Zhiyong Wu 0003
  • Weizhong Zhang
  • Xiaodan Liang

There is a rising interest in further exploring the zero-shot learning potential of large pre-trained language models (PLMs). A new paradigm called data-generation-based zero-shot learning has achieved impressive success. In this paradigm, the synthesized data from the PLM acts as the carrier of knowledge, which is used to train a task-specific model with orders of magnitude fewer parameters than the PLM, achieving both higher performance and efficiency than prompt-based zero-shot learning methods on PLMs. The main hurdle of this approach is that the synthesized data from PLM usually contains a significant portion of low-quality samples. Fitting on such data will greatly hamper the performance of the task-specific model, making it unreliable for deployment. Previous methods remedy this issue mainly by filtering synthetic data using heuristic metrics(e.g., output confidence), or refining the data with the help of a human expert, which comes with excessive manual tuning or expensive costs. In this paper, we propose a novel noise-robust re-weighting framework SunGen to automatically construct high-quality data for zero-shot classification problems. Our framework features the ability to learn the sample weights indicating data quality without requiring any human annotation. We theoretically and empirically verify the ability of our method to help construct good-quality synthetic datasets. Notably, SunGen-LSTM yields a 9.8% relative improvement than the baseline on average accuracy across eight different established text classification tasks.

Details

NeurIPS Conference 2023 Conference Paper

T2I-CompBench: A Comprehensive Benchmark for Open-world Compositional Text-to-image Generation

  • Kaiyi Huang
  • Kaiyue Sun
  • Enze Xie
  • Zhenguo Li
  • Xihui Liu

Despite the stunning ability to generate high-quality images by recent text-to-image models, current approaches often struggle to effectively compose objects with different attributes and relationships into a complex and coherent scene. We propose T2I-CompBench, a comprehensive benchmark for open-world compositional text-to-image generation, consisting of 6, 000 compositional text prompts from 3 categories (attribute binding, object relationships, and complex compositions) and 6 sub-categories (color binding, shape binding, texture binding, spatial relationships, non-spatial relationships, and complex compositions). We further propose several evaluation metrics specifically designed to evaluate compositional text-to-image generation and explore the potential and limitations of multimodal LLMs for evaluation. We introduce a new approach, Generative mOdel finetuning with Reward-driven Sample selection (GORS), to boost the compositional text-to-image generation abilities of pretrained text-to-image models. Extensive experiments and evaluations are conducted to benchmark previous methods on T2I-CompBench, and to validate the effectiveness of our proposed evaluation metrics and GORS approach. Project page is available at https: //karine-h. github. io/T2I-CompBench/.

PDF Details

ICLR Conference 2023 Conference Paper

Task-customized Masked Autoencoder via Mixture of Cluster-conditional Experts

  • Zhili Liu
  • Kai Chen 0023
  • Jianhua Han
  • Lanqing Hong
  • Hang Xu 0004
  • Zhenguo Li
  • James T. Kwok

Masked Autoencoder (MAE) is a prevailing self-supervised learning method that achieves promising results in model pre-training. However, when the various downstream tasks have data distributions different from the pre-training data, the semantically irrelevant pre-training information might result in negative transfer, impeding MAE’s scalability. To address this issue, we propose a novel MAE-based pre-training paradigm, Mixture of Cluster-conditional Experts (MoCE), which can be trained once but provides customized pre-training models for diverse downstream tasks. Different from the mixture of experts (MoE), our MoCE trains each expert only with semantically relevant images by using cluster-conditional gates. Thus, each downstream task can be allocated to its customized model pre-trained with data most similar to the downstream data. Experiments on a collection of 11 downstream tasks show that MoCE outperforms the vanilla MAE by 2.45\% on average. It also obtains new state-of-the-art self-supervised learning results on detection and segmentation.

Details

AAAI Conference 2022 Conference Paper

AutoBERT-Zero: Evolving BERT Backbone from Scratch

  • Jiahui Gao
  • Hang Xu
  • Han Shi
  • Xiaozhe Ren
  • Philip L. H. Yu
  • Xiaodan Liang
  • Xin Jiang
  • Zhenguo Li

Transformer-based pre-trained language models like BERT and its variants have recently achieved promising performance in various natural language processing (NLP) tasks. However, the conventional paradigm constructs the backbone by purely stacking the manually designed global selfattention layers, introducing inductive bias and thus leads to sub-optimal. In this work, we make the first attempt to automatically discover novel pre-trained language model (PLM) backbone on a flexible search space containing the most fundamental operations from scratch. Specifically, we propose a well-designed search space which (i) contains primitive math operations in the intra-layer level to explore novel attention structures, and (ii) leverages convolution blocks to be the supplementary for attentions in the inter-layer level to better learn local dependency. To enhance the efficiency for finding promising architectures, we propose an Operation- Priority Neural Architecture Search (OP-NAS) algorithm, which optimizes both the search algorithm and evaluation of candidate models. Specifically, we propose Operation- Priority (OP) evolution strategy to facilitate model search via balancing exploration and exploitation. Furthermore, we design a Bi-branch Weight-Sharing (BIWS) training strategy for fast model evaluation. Extensive experiments show that the searched architecture (named AutoBERT-Zero) significantly outperforms BERT and its variants of different model capacities in various downstream tasks, proving the architecture’s transfer and scaling abilities. Remarkably, AutoBERT-Zerobase outperforms RoBERTa-base (using much more data) and BERT-large (with much larger model size) by 2. 4 and 1. 4 higher score on GLUE test set.

PDF Details

NeurIPS Conference 2022 Conference Paper

CAGroup3D: Class-Aware Grouping for 3D Object Detection on Point Clouds

  • Haiyang Wang
  • Lihe Ding
  • Shaocong Dong
  • Shaoshuai Shi
  • Aoxue Li
  • Jianan Li
  • Zhenguo Li
  • Liwei Wang

We present a novel two-stage fully sparse convolutional 3D object detection framework, named CAGroup3D. Our proposed method first generates some high-quality 3D proposals by leveraging the class-aware local group strategy on the object surface voxels with the same semantic predictions, which considers semantic consistency and diverse locality abandoned in previous bottom-up approaches. Then, to recover the features of missed voxels due to incorrect voxel-wise segmentation, we build a fully sparse convolutional RoI pooling module to directly aggregate fine-grained spatial information from backbone for further proposal refinement. It is memory-and-computation efficient and can better encode the geometry-specific features of each 3D proposal. Our model achieves state-of-the-art 3D detection performance with remarkable gains of +3. 6% on ScanNet V2 and +2. 6% on SUN RGB-D in term of mAP@0. 25. Code will be available at https: //github. com/Haiyang-W/CAGroup3D.

PDF Details

NeurIPS Conference 2022 Conference Paper

DetCLIP: Dictionary-Enriched Visual-Concept Paralleled Pre-training for Open-world Detection

  • Lewei Yao
  • Jianhua Han
  • Youpeng Wen
  • Xiaodan Liang
  • Dan Xu
  • Wei Zhang
  • Zhenguo Li
  • Chunjing Xu

Open-world object detection, as a more general and challenging goal, aims to recognize and localize objects described by arbitrary category names. The recent work GLIP formulates this problem as a grounding problem by concatenating all category names of detection datasets into sentences, which leads to inefficient interaction between category names. This paper presents DetCLIP, a paralleled visual-concept pre-training method for open-world detection by resorting to knowledge enrichment from a designed concept dictionary. To achieve better learning efficiency, we propose a novel paralleled concept formulation that extracts concepts separately to better utilize heterogeneous datasets (i. e. , detection, grounding, and image-text pairs) for training. We further design a concept dictionary (with descriptions) from various online sources and detection datasets to provide prior knowledge for each concept. By enriching the concepts with their descriptions, we explicitly build the relationships among various concepts to facilitate the open-domain learning. The proposed concept dictionary is further used to provide sufficient negative concepts for the construction of the word-region alignment loss, and to complete labels for objects with missing descriptions in captions of image-text pair data. The proposed framework demonstrates strong zero-shot detection performances, e. g. , on the LVIS dataset, our DetCLIP-T outperforms GLIP-T by 9. 9% mAP and obtains a 13. 5% improvement on rare categories compared to the fully-supervised model with the same backbone as ours.

PDF Details

TMLR Journal 2022 Journal Article

DHA: End-to-End Joint Optimization of Data Augmentation Policy, Hyper-parameter and Architecture

  • Kaichen Zhou
  • Lanqing Hong
  • Shoukang Hu
  • Fengwei Zhou
  • Binxin Ru
  • Jiashi Feng
  • Zhenguo Li

Automated machine learning (AutoML) usually involves several crucial components, such as Data Augmentation (DA) policy, Hyper-Parameter Optimization (HPO), and Neural Architecture Search (NAS). Although many strategies have been developed for automating these components in separation, joint optimization of these components remains challenging due to the largely increased search dimension and the variant input types of each component. In parallel to this, the common practice of searching for the optimal architecture first and then retraining it before deployment in NAS often suffers from the low-performance correlation between the searching and retraining stages. An end-to-end solution that integrates the AutoML components and returns a ready-to-use model at the end of the search is desirable. In view of these, we propose DHA, which achieves joint optimization of Data augmentation policy, Hyper-parameter, and Architecture. Specifically, end-to-end NAS is achieved in a differentiable manner by optimizing a compressed lower-dimensional feature space, while DA policy and HPO are regarded as dynamic schedulers, which adapt themselves to the update of network parameters and network architecture at the same time. Experiments show that DHA achieves state-of-the-art (SOTA) results on various datasets and search spaces. To the best of our knowledge, we are the first to efficiently and jointly optimize DA policy, NAS, and HPO in an end-to-end manner without retraining.

PDF Details

ICLR Conference 2022 Conference Paper

FILIP: Fine-grained Interactive Language-Image Pre-Training

  • Lewei Yao
  • Runhui Huang
  • Lu Hou 0002
  • Guansong Lu
  • Minzhe Niu
  • Hang Xu 0004
  • Xiaodan Liang
  • Zhenguo Li

Unsupervised large-scale vision-language pre-training has shown promising advances on various downstream tasks. Existing methods often model the cross-modal interaction either via the similarity of the global feature of each modality which misses sufficient information, or finer-grained interactions using cross/self-attention upon visual and textual tokens. However, cross/self-attention suffers from inferior efficiency in both training and inference. In this paper, we introduce a large-scale Fine-grained Interactive Language-Image Pre-training (FILIP) to achieve finer-level alignment through a cross-modal late interaction mechanism, which uses a token-wise maximum similarity between visual and textual tokens to guide the contrastive objective. FILIP successfully leverages the finer-grained expressiveness between image patches and textual words by modifying only contrastive loss, while simultaneously gaining the ability to pre-compute image and text representations offline at inference, keeping both large-scale training and inference efficient. Furthermore, we construct a new large-scale image-text pair dataset called FILIP300M for pre-training. Experiments show that FILIP achieves state-of-the-art performance on multiple downstream vision-language tasks including zero-shot image classification and image-text retrieval. The visualization on word-patch alignment further shows that FILIP can learn meaningful fine-grained features with promising localization ability.

Details

ICLR Conference 2022 Conference Paper

Generalizing Few-Shot NAS with Gradient Matching

  • Shoukang Hu
  • Ruochen Wang
  • Lanqing Hong
  • Zhenguo Li
  • Cho-Jui Hsieh
  • Jiashi Feng

Efficient performance estimation of architectures drawn from large search spaces is essential to Neural Architecture Search. One-Shot methods tackle this challenge by training one supernet to approximate the performance of every architecture in the search space via weight-sharing, thereby drastically reducing the search cost. However, due to coupled optimization between child architectures caused by weight-sharing, One-Shot supernet's performance estimation could be inaccurate, leading to degraded search outcomes. To address this issue, Few-Shot NAS reduces the level of weight-sharing by splitting the One-Shot supernet into multiple separated sub-supernets via edge-wise (layer-wise) exhaustive partitioning. Since each partition of the supernet is not equally important, it necessitates the design of a more effective splitting criterion. In this work, we propose a gradient matching score (GM) that leverages gradient information at the shared weight for making informed splitting decisions. Intuitively, gradients from different child models can be used to identify whether they agree on how to update the shared modules, and subsequently to decide if they should share weight. Compared with exhaustive partitioning, the proposed criterion significantly reduces the branching factor per edge. This allows us to split more edges (layers) for a given budget, resulting in substantially improved performance as NAS search spaces usually include dozens of edges (layers). Extensive empirical evaluations of the proposed method on a wide range of search spaces (NASBench-201, DARTS, MobileNet Space), datasets (cifar10, cifar100, ImageNet) and search algorithms (DARTS, SNAS, RSPS, ProxylessNAS, OFA) demonstrate that it significantly outperforms its Few-Shot counterparts while surpassing previous comparable methods in terms of the accuracy of derived architectures. Our code is available at https://github.com/skhu101/GM-NAS.

Details

ICLR Conference 2022 Conference Paper

How Well Does Self-Supervised Pre-Training Perform with Streaming Data?

  • Dapeng Hu
  • Shipeng Yan
  • Qizhengqiu Lu
  • Lanqing Hong
  • Hailin Hu 0002
  • Yifan Zhang 0004
  • Zhenguo Li
  • Xinchao Wang

Prior works on self-supervised pre-training focus on the joint training scenario, where massive unlabeled data are assumed to be given as input all at once, and only then is a learner trained. Unfortunately, such a problem setting is often impractical if not infeasible since many real-world tasks rely on sequential learning, e.g., data are decentralized or collected in a streaming fashion. In this paper, we conduct the first thorough and dedicated investigation on self-supervised pre-training with streaming data, aiming to shed light on the model behavior under this overlooked setup. Specifically, we pre-train over 500 models on four categories of pre-training streaming data from ImageNet and DomainNet and evaluate them on three types of downstream tasks and 12 different downstream datasets. Our studies show that, somehow beyond our expectation, with simple data replay or parameter regularization, sequential self-supervised pre-training turns out to be an efficient alternative for joint pre-training, as the performances of the former are mostly on par with those of the latter. Moreover, catastrophic forgetting, a common issue in sequential supervised learning, is much alleviated in sequential self-supervised learning (SSL), which is well justified through our comprehensive empirical analysis on representations and the sharpness of minima in the loss landscape. Our findings, therefore, suggest that, in practice, for SSL, the cumbersome joint training can be replaced mainly by sequential learning, which in turn enables a much broader spectrum of potential application scenarios.

Details

ICLR Conference 2022 Conference Paper

Memory Replay with Data Compression for Continual Learning

  • Liyuan Wang
  • Xingxing Zhang 0001
  • Kuo Yang
  • Longhui Yu
  • Chongxuan Li
  • Lanqing Hong
  • Shifeng Zhang
  • Zhenguo Li

Continual learning needs to overcome catastrophic forgetting of the past. Memory replay of representative old training samples has been shown as an effective solution, and achieves the state-of-the-art (SOTA) performance. However, existing work is mainly built on a small memory buffer containing a few original data, which cannot fully characterize the old data distribution. In this work, we propose memory replay with data compression to reduce the storage cost of old training samples and thus increase their amount that can be stored in the memory buffer. Observing that the trade-off between the quality and quantity of compressed data is highly nontrivial for the efficacy of memory replay, we propose a novel method based on determinantal point processes (DPPs) to efficiently determine an appropriate compression quality for currently-arrived training samples. In this way, using a naive data compression algorithm with a properly selected quality can largely boost recent strong baselines by saving more compressed data in a limited storage space. We extensively validate this across several benchmarks of class-incremental learning and in a realistic scenario of object detection for autonomous driving.

Details

ICLR Conference 2022 Conference Paper

Nonlinear ICA Using Volume-Preserving Transformations

  • Xiaojiang Yang
  • Yi Wang
  • Jiacheng Sun
  • Xing Zhang
  • Shifeng Zhang
  • Zhenguo Li
  • Junchi Yan

Nonlinear ICA is a fundamental problem in machine learning, aiming to identify the underlying independent components (sources) from data which is assumed to be a nonlinear function (mixing function) of these sources. Recent works prove that if the sources have some particular structures (e.g. temporal structure), they are theoretically identifiable even if the mixing function is arbitrary. However, in many cases such restrictions on the sources are difficult to satisfy or even verify, hence it inhibits the applicability of the proposed methods. Different from these works, we propose a general framework for nonlinear ICA, in which the mixing function is assumed to be a volume-preserving transformation, and meanwhile the conditions on the sources can be much looser. We provide an insightful proof of the identifiability of the proposed framework. We implement the framework by volume-preserving Flow-based models, and verify our theory by experiments on artificial data and synthesized images. Moreover, results on real-world images indicate that our framework can disentangle interpretable features.

Details

ICLR Conference 2022 Conference Paper

On Redundancy and Diversity in Cell-based Neural Architecture Search

  • Xingchen Wan
  • Binxin Ru
  • Pedro M. Esperança
  • Zhenguo Li

Searching for the architecture cells is a dominant paradigm in NAS. However, little attention has been devoted to the analysis of the cell-based search spaces even though it is highly important for the continual development of NAS. In this work, we conduct an empirical post-hoc analysis of architectures from the popular cell-based search spaces and find that the existing search spaces contain a high degree of redundancy: the architecture performance is less sensitive to changes at large parts of the cells, and universally adopted design rules, like the explicit search for a reduction cell, significantly increase the complexities but have very limited impact on the performance. Across architectures found by a diverse set of search strategies, we consistently find that the parts of the cells that do matter for architecture performance often follow similar and simple patterns. By constraining cells to include these patterns, randomly sampled architectures can match or even outperform the state of the art. These findings cast doubts into our ability to discover truly novel architectures in the existing cell-based search spaces and, inspire our suggestions for improvement to guide future NAS research. Code is available at https://github.com/xingchenwan/cell-based-NAS-analysis.

Details

ICLR Conference 2022 Conference Paper

Rethinking Adversarial Transferability from a Data Distribution Perspective

  • Yao Zhu
  • Jiacheng Sun
  • Zhenguo Li

Adversarial transferability enables attackers to generate adversarial examples from the source model to attack the target model, which has raised security concerns about the deployment of DNNs in practice. In this paper, we rethink adversarial transferability from a data distribution perspective and further enhance transferability by score matching based optimization. We identify that some samples with injecting small Gaussian noise can fool different target models, and their adversarial examples under different source models have much stronger transferability. We hypothesize that these samples are in the low-density region of the ground truth distribution where models are not well trained. To improve the attack success rate of adversarial examples, we match the adversarial attacks with the directions which effectively decrease the ground truth density. We propose Intrinsic Adversarial Attack (IAA), which smooths the activation function and decreases the impact of the later layers of a given normal model, to increase the alignment of adversarial attack and the gradient of joint data distribution. We conduct comprehensive transferable attacks against multiple DNNs and show that our IAA can boost the transferability of the crafted attacks in all cases and go beyond state-of-the-art methods.

Details

ICLR Conference 2022 Conference Paper

Revisiting Over-smoothing in BERT from the Perspective of Graph

  • Han Shi
  • Jiahui Gao
  • Hang Xu 0004
  • Xiaodan Liang
  • Zhenguo Li
  • Lingpeng Kong
  • Stephen M. S. Lee
  • James T. Kwok

Recently over-smoothing phenomenon of Transformer-based models is observed in both vision and language fields. However, no existing work has delved deeper to further investigate the main cause of this phenomenon. In this work, we make the attempt to analyze the over-smoothing problem from the perspective of graph, where such problem was first discovered and explored. Intuitively, the self-attention matrix can be seen as a normalized adjacent matrix of a corresponding graph. Based on the above connection, we provide some theoretical analysis and find that layer normalization plays a key role in the over-smoothing issue of Transformer-based models. Specifically, if the standard deviation of layer normalization is sufficiently large, the output of Transformer stacks will converge to a specific low-rank subspace and result in over-smoothing. To alleviate the over-smoothing problem, we consider hierarchical fusion strategies, which combine the representations from different layers adaptively to make the output more diverse. Extensive experiment results on various data sets illustrate the effect of our fusion method.

Details

AAAI Conference 2022 Conference Paper

Task-Customized Self-Supervised Pre-training with Scalable Dynamic Routing

  • Zhili LIU
  • Jianhua Han
  • Lanqing Hong
  • Hang Xu
  • Kai Chen
  • Chunjing Xu
  • Zhenguo Li

Self-supervised learning (SSL), especially contrastive methods, has raised attraction recently as it learns effective transferable representations without semantic annotations. A common practice for self-supervised pre-training is to use as much data as possible. For a specific downstream task, however, involving irrelevant data in pre-training may degenerate the downstream performance, observed from our extensive experiments. On the other hand, for existing SSL methods, it is burdensome and infeasible to use different downstreamtask-customized datasets in pre-training for different tasks. To address this issue, we propose a novel SSL paradigm called Scalable Dynamic Routing (SDR), which can be trained once and deployed efficiently to different downstream tasks with task-customized pre-trained models. Specifically, we construct the SDRnet with various sub-nets and train each sub-net with only one subset of the data by data-aware progressive training. When a downstream task arrives, we route among all the pre-trained sub-nets to get the best along with its corresponding weights. Experiment results show that our SDR can train 256 sub-nets on ImageNet simultaneously, which provides better transfer performance than a unified model trained on the full ImageNet, achieving state-of-the-art (SOTA) averaged accuracy over 11 downstream classification tasks and AP on PASCAL VOC detection task.

PDF Details

NeurIPS Conference 2022 Conference Paper

Understanding Square Loss in Training Overparametrized Neural Network Classifiers

  • Tianyang Hu
  • Jun Wang
  • Wenjia Wang
  • Zhenguo Li

Deep learning has achieved many breakthroughs in modern classification tasks. Numerous architectures have been proposed for different data structures but when it comes to the loss function, the cross-entropy loss is the predominant choice. Recently, several alternative losses have seen revived interests for deep classifiers. In particular, empirical evidence seems to promote square loss but a theoretical justification is still lacking. In this work, we contribute to the theoretical understanding of square loss in classification by systematically investigating how it performs for overparametrized neural networks in the neural tangent kernel (NTK) regime. Interesting properties regarding the generalization error, robustness, and calibration error are revealed. We consider two cases, according to whether classes are separable or not. In the general non-separable case, fast convergence rate is established for both misclassification rate and calibration error. When classes are separable, the misclassification rate improves to be exponentially fast. Further, the resulting margin is proven to be lower bounded away from zero, providing theoretical guarantees for robustness. We expect our findings to hold beyond the NTK regime and translate to practical settings. To this end, we conduct extensive empirical studies on practical neural networks, demonstrating the effectiveness of square loss in both synthetic low-dimensional data and real image data. Comparing to cross-entropy, square loss has comparable generalization error but noticeable advantages in robustness and model calibration.

PDF Details

NeurIPS Conference 2022 Conference Paper

ZooD: Exploiting Model Zoo for Out-of-Distribution Generalization

  • Qishi Dong
  • Awais Muhammad
  • Fengwei Zhou
  • Chuanlong Xie
  • Tianyang Hu
  • Yongxin Yang
  • Sung-Ho Bae
  • Zhenguo Li

Recent advances on large-scale pre-training have shown great potentials of leveraging a large set of Pre-Trained Models (PTMs) for improving Out-of-Distribution (OoD) generalization, for which the goal is to perform well on possible unseen domains after fine-tuning on multiple training domains. However, maximally exploiting a zoo of PTMs is challenging since fine-tuning all possible combinations of PTMs is computationally prohibitive while accurate selection of PTMs requires tackling the possible data distribution shift for OoD tasks. In this work, we propose ZooD, a paradigm for PTMs ranking and ensemble with feature selection. Our proposed metric ranks PTMs by quantifying inter-class discriminability and inter-domain stability of the features extracted by the PTMs in a leave-one-domain-out cross-validation manner. The top-K ranked models are then aggregated for the target OoD task. To avoid accumulating noise induced by model ensemble, we propose an efficient variational EM algorithm to select informative features. We evaluate our paradigm on a diverse model zoo consisting of 35 models for various OoD tasks and demonstrate: (i) model ranking is better correlated with fine-tuning ranking than previous methods and up to 9859x faster than brute-force fine-tuning; (ii) OoD generalization after model ensemble with feature selection outperforms the state-of-the-art methods and the accuracy on most challenging task DomainNet is improved from 46. 5\% to 50. 6\%. Furthermore, we provide the fine-tuning results of 35 PTMs on 7 OoD datasets, hoping to help the research of model zoo and OoD generalization. Code will be available at \href{https: //gitee. com/mindspore/models/tree/master/research/cv/zood}{https: //gitee. com/mindspore/models/tree/master/research/cv/zood}.

PDF Details

AAAI Conference 2021 Conference Paper

Ada-Segment: Automated Multi-loss Adaptation for Panoptic Segmentation

  • Gengwei Zhang
  • Yiming Gao
  • Hang Xu
  • Hao Zhang
  • Zhenguo Li
  • Xiaodan Liang

Panoptic segmentation that unifies instance segmentation and semantic segmentation has recently attracted increasing attention. While most existing methods focus on designing novel architectures, we steer toward a different perspective: performing automated multi-loss adaptation (named Ada- Segment) on the fly to flexibly adjust multiple training losses over the course of training using a controller trained to capture the learning dynamics. This offers a few advantages: it bypasses manual tuning of the sensitive loss combination, a decisive factor for panoptic segmentation; it allows to explicitly model the learning dynamics, and reconcile the learning of multiple objectives (up to ten in our experiments); with an end-to-end architecture, it generalizes to different datasets without the need of re-tuning hyperparameters or readjusting the training process laboriously. Our Ada-Segment brings 2. 7% panoptic quality (PQ) improvement on COCO val split from the vanilla baseline, achieving the state-of-theart 48. 5% PQ on COCO test-dev split and 32. 9% PQ on ADE20K dataset. The extensive ablation studies reveal the ever-changing dynamics throughout the training process, necessitating the incorporation of an automated and adaptive learning strategy as presented in this paper.

PDF Details

AAAI Conference 2021 Conference Paper

DecAug: Out-of-Distribution Generalization via Decomposed Feature Representation and Semantic Augmentation

  • Haoyue Bai
  • Rui Sun
  • Lanqing Hong
  • Fengwei Zhou
  • Nanyang Ye
  • Han-Jia Ye
  • S.-H. Gary Chan
  • Zhenguo Li

While deep learning demonstrates its strong ability to handle independent and identically distributed (IID) data, it often suffers from out-of-distribution (OoD) generalization, where the test data come from another distribution (w. r. t. the training one). Designing a general OoD generalization framework for a wide range of applications is challenging, mainly due to different kinds of distribution shifts in the real world, such as the shift across domains or the extrapolation of correlation. Most of the previous approaches can only solve one specific distribution shift, leading to unsatisfactory performance when applied to various OoD benchmarks. In this work, we propose DecAug, a novel decomposed feature representation and semantic augmentation approach for OoD generalization. Specifically, DecAug disentangles the category-related and context-related features by orthogonalizing the two gradients (w. r. t. intermediate features) of losses for predicting category and context labels, where category-related features contain causal information of the target object, while contextrelated features cause distribution shifts between training and test data. Furthermore, we perform gradient-based augmentation on context-related features to improve the robustness of learned representations. Experimental results show that DecAug outperforms other state-of-the-art methods on various OoD datasets, which is among the very few methods that can deal with different types of OoD generalization challenges.

PDF Details

AAAI Conference 2021 Conference Paper

How to Save your Annotation Cost for Panoptic Segmentation?

  • Xuefeng Du
  • ChenHan Jiang
  • Hang Xu
  • Gengwei Zhang
  • Zhenguo Li

How to properly reduce the annotation cost for panoptic segmentation? How to leverage and optimize the cost-quality trade-off for training data and model? These questions are key challenges towards a label-efficient and scalable panoptic segmentation system due to its expensive instance/semantic pixel-level annotation requirements. By closely examining different kinds of cheaper labels, we introduce a novel multiobjective framework to automatically determine the allocation of different annotations, so as to reach a better segmentation quality with a lower annotation cost. Specifically, we design a Cost-Quality Balanced Network (CQB-Net) to generate the panoptic segmentation map, which distills the crucial relations between various supervisions including panoptic labels, image-level classification labels, bounding boxes, and the semantic coherence information between the foreground and background. Instead of ad-hoc allocation during training, we formulate the optimization of cost-quality trade-off as a Multi-Objective Optimization Problem (MOOP). We model the marginal quality improvement of each annotation and approximate the Pareto-front to enable a label-efficient allocation ratio. Extensive experiments on COCO benchmark show the superiority of our method, e. g. achieving a segmentation quality of 43. 4% compared to 43. 0% of OCFusion while saving 2. 4x annotation cost.

PDF Details

NeurIPS Conference 2021 Conference Paper

iFlow: Numerically Invertible Flows for Efficient Lossless Compression via a Uniform Coder

  • Shifeng Zhang
  • Ning Kang
  • Tom Ryder
  • Zhenguo Li

It was estimated that the world produced $59 ZB$ ($5. 9 \times 10^{13} GB$) of data in 2020, resulting in the enormous costs of both data storage and transmission. Fortunately, recent advances in deep generative models have spearheaded a new class of so-called "neural compression" algorithms, which significantly outperform traditional codecs in terms of compression ratio. Unfortunately, the application of neural compression garners little commercial interest due to its limited bandwidth; therefore, developing highly efficient frameworks is of critical practical importance. In this paper, we discuss lossless compression using normalizing flows which have demonstrated a great capacity for achieving high compression ratios. As such, we introduce iFlow, a new method for achieving efficient lossless compression. We first propose Modular Scale Transform (MST) and a novel family of numerically invertible flow transformations based on MST. Then we introduce the Uniform Base Conversion System (UBCS), a fast uniform-distribution codec incorporated into iFlow, enabling efficient compression. iFlow achieves state-of-the-art compression ratios and is $5 \times$ quicker than other high-performance schemes. Furthermore, the techniques presented in this paper can be used to accelerate coding time for a broad class of flow-based algorithms.

PDF Details

ICLR Conference 2021 Conference Paper

Loss Function Discovery for Object Detection via Convergence-Simulation Driven Search

  • Peidong Liu
  • Gengwei Zhang
  • Bochao Wang
  • Hang Xu 0004
  • Xiaodan Liang
  • Yong Jiang
  • Zhenguo Li

Designing proper loss functions for vision tasks has been a long-standing research direction to advance the capability of existing models. For object detection, the well-established classification and regression loss functions have been carefully designed by considering diverse learning challenges (e.g. class imbalance, hard negative samples, and scale variances). Inspired by the recent progress in network architecture search, it is interesting to explore the possibility of discovering new loss function formulations via directly searching the primitive operation combinations. So that the learned losses not only fit for diverse object detection challenges to alleviate huge human efforts, but also have better alignment with evaluation metric and good mathematical convergence property. Beyond the previous auto-loss works on face recognition and image classification, our work makes the first attempt to discover new loss functions for the challenging object detection from primitive operation levels and finds the searched losses are insightful. We propose an effective convergence-simulation driven evolutionary search algorithm, called CSE-Autoloss, for speeding up the search progress by regularizing the mathematical rationality of loss candidates via two progressive convergence simulation modules: convergence property verification and model optimization simulation. CSE-Autoloss involves the search space (i.e. 21 mathematical operators, 3 constant-type inputs, and 3 variable-type inputs) that cover a wide range of the possible variants of existing losses and discovers best-searched loss function combination within a short time (around 1.5 wall-clock days with 20x speedup in comparison to the vanilla evolutionary algorithm). We conduct extensive evaluations of loss function search on popular detectors and validate the good generalization capability of searched losses across diverse architectures and various datasets. Our experiments show that the best-discovered loss function combinations outperform default combinations (Cross-entropy/Focal loss for classification and L1 loss for regression) by 1.1% and 0.8% in terms of mAP for two-stage and one-stage detectors on COCO respectively. Our searched losses are available at https://github.com/PerdonLiu/CSE-Autoloss.

Details

AAAI Conference 2021 Conference Paper

MetaAugment: Sample-Aware Data Augmentation Policy Learning

  • Fengwei Zhou
  • Jiawei Li
  • Chuanlong Xie
  • Fei Chen
  • Lanqing Hong
  • Rui Sun
  • Zhenguo Li

Automated data augmentation has shown superior performance in image recognition. Existing works search for datasetlevel augmentation policies without considering individual sample variations, which are likely to be sub-optimal. On the other hand, learning different policies for different samples naively could greatly increase the computing cost. In this paper, we learn a sample-aware data augmentation policy efficiently by formulating it as a sample reweighting problem. Specifically, an augmentation policy network takes a transformation and the corresponding augmented image as inputs, and outputs a weight to adjust the augmented image loss computed by a task network. At training stage, the task network minimizes the weighted losses of augmented training images, while the policy network minimizes the loss of the task network on a validation set via meta-learning. We theoretically prove the convergence of the training procedure and further derive the exact convergence rate. Superior performance is achieved on widely-used benchmarks including CIFAR-10/100, Omniglot, and ImageNet.

PDF Details

NeurIPS Conference 2021 Conference Paper

MixACM: Mixup-Based Robustness Transfer via Distillation of Activated Channel Maps

  • Awais Muhammad
  • Fengwei Zhou
  • Chuanlong Xie
  • Jiawei Li
  • Sung-Ho Bae
  • Zhenguo Li

Deep neural networks are susceptible to adversarially crafted, small, and imperceptible changes in the natural inputs. The most effective defense mechanism against these examples is adversarial training which constructs adversarial examples during training by iterative maximization of loss. The model is then trained to minimize the loss on these constructed examples. This min-max optimization requires more data, larger capacity models, and additional computing resources. It also degrades the standard generalization performance of a model. Can we achieve robustness more efficiently? In this work, we explore this question from the perspective of knowledge transfer. First, we theoretically show the transferability of robustness from an adversarially trained teacher model to a student model with the help of mixup augmentation. Second, we propose a novel robustness transfer method called Mixup-Based Activated Channel Maps (MixACM) Transfer. MixACM transfers robustness from a robust teacher to a student by matching activated channel maps generated without expensive adversarial perturbations. Finally, extensive experiments on multiple datasets and different learning scenarios show our method can transfer robustness while also improving generalization on natural images.

PDF Details

NeurIPS Conference 2021 Conference Paper

On Effective Scheduling of Model-based Reinforcement Learning

  • Hang Lai
  • Jian Shen
  • Weinan Zhang
  • Yimin Huang
  • Xing Zhang
  • Ruiming Tang
  • Yong Yu
  • Zhenguo Li

Model-based reinforcement learning has attracted wide attention due to its superior sample efficiency. Despite its impressive success so far, it is still unclear how to appropriately schedule the important hyperparameters to achieve adequate performance, such as the real data ratio for policy optimization in Dyna-style model-based algorithms. In this paper, we first theoretically analyze the role of real data in policy training, which suggests that gradually increasing the ratio of real data yields better performance. Inspired by the analysis, we propose a framework named AutoMBPO to automatically schedule the real data ratio as well as other hyperparameters in training model-based policy optimization (MBPO) algorithm, a representative running case of model-based methods. On several continuous control tasks, the MBPO instance trained with hyperparameters scheduled by AutoMBPO can significantly surpass the original one, and the real data ratio schedule found by AutoMBPO shows consistency with our theoretical analysis.

PDF Details

NeurIPS Conference 2021 Conference Paper

One Million Scenes for Autonomous Driving: ONCE Dataset

  • Jiageng Mao
  • Niu Minzhe
  • ChenHan Jiang
  • hanxue liang
  • Jingheng Chen
  • Xiaodan Liang
  • Yamin Li
  • Chaoqiang Ye

Current perception models in autonomous driving have become notorious for greatly relying on a mass of annotated data to cover unseen cases and address the long-tail problem. On the other hand, learning from unlabeled large-scale collected data and incrementally self-training powerful recognition models have received increasing attention and may become the solutions of next-generation industry-level powerful and robust perception models in autonomous driving. However, the research community generally suffered from data inadequacy of those essential real-world scene data, which hampers the future exploration of fully/semi/self-supervised methods for 3D perception. In this paper, we introduce the ONCE (One millioN sCenEs) dataset for 3D object detection in the autonomous driving scenario. The ONCE dataset consists of 1 million LiDAR scenes and 7 million corresponding camera images. The data is selected from 144 driving hours, which is 20x longer than the largest 3D autonomous driving dataset available (\eg nuScenes and Waymo), and it is collected across a range of different areas, periods and weather conditions. To facilitate future research on exploiting unlabeled data for 3D detection, we additionally provide a benchmark in which we reproduce and evaluate a variety of self-supervised and semi-supervised methods on the ONCE dataset. We conduct extensive analyses on those methods and provide valuable observations on their performance related to the scale of used data. Data, code, and more information are available at \href{https: //once-for-auto-driving. github. io/index. html}{http: //www. once-for-auto-driving. com}.

PDF Details

NeurIPS Conference 2021 Conference Paper

OSOA: One-Shot Online Adaptation of Deep Generative Models for Lossless Compression

  • Chen Zhang
  • Shifeng Zhang
  • Fabio Maria Carlucci
  • Zhenguo Li

Explicit deep generative models (DGMs), e. g. , VAEs and Normalizing Flows, have shown to offer an effective data modelling alternative for lossless compression. However, DGMs themselves normally require large storage space and thus contaminate the advantage brought by accurate data density estimation. To eliminate the requirement of saving separate models for different target datasets, we propose a novel setting that starts from a pretrained deep generative model and compresses the data batches while adapting the model with a dynamical system for only one epoch. We formalise this setting as that of One-Shot Online Adaptation (OSOA) of DGMs for lossless compression and propose a vanilla algorithm under this setting. Experimental results show that vanilla OSOA can save significant time versus training bespoke models and space versus using one model for all targets. With the same adaptation step number or adaptation time, it is shown vanilla OSOA can exhibit better space efficiency, e. g. , $47\%$ less space, than fine-tuning the pretrained model and saving the fine-tuned model. Moreover, we showcase the potential of OSOA and motivate more sophisticated OSOA algorithms by showing further space or time efficiency with multiple updates per batch and early stopping.

PDF Details

NeurIPS Conference 2021 Conference Paper

SODA10M: A Large-Scale 2D Self/Semi-Supervised Object Detection Dataset for Autonomous Driving

  • Jianhua Han
  • Xiwen Liang
  • Hang Xu
  • Kai Chen
  • Lanqing Hong
  • Jiageng Mao
  • Chaoqiang Ye
  • Wei Zhang

Aiming at facilitating a real-world, ever-evolving and scalable autonomous driving system, we present a large-scale dataset for standardizing the evaluation of different self-supervised and semi-supervised approaches by learning from raw data, which is the first and largest dataset to date. Existing autonomous driving systems heavily rely on `perfect' visual perception models (i. e. , detection) trained using extensive annotated data to ensure safety. However, it is unrealistic to elaborately label instances of all scenarios and circumstances (i. e. , night, extreme weather, cities) when deploying a robust autonomous driving system. Motivated by recent advances of self-supervised and semi-supervised learning, a promising direction is to learn a robust detection model by collaboratively exploiting large-scale unlabeled data and few labeled data. Existing datasets (i. e. , BDD100K, Waymo) either provide only a small amount of data or covers limited domains with full annotation, hindering the exploration of large-scale pre-trained models. Here, we release a Large-Scale 2D Self/semi-supervised Object Detection dataset for Autonomous driving, named as SODA10M, containing 10 million unlabeled images and 20K images labeled with 6 representative object categories. To improve diversity, the images are collected within 27833 driving hours under different weather conditions, periods and location scenes of 32 different cities. We provide extensive experiments and deep analyses of existing popular self-supervised and semi-supervised approaches, and some interesting findings in autonomous driving scope. Experiments show that SODA10M can serve as a promising pre-training dataset for different self-supervised learning methods, which gives superior performance when finetuning with different downstream tasks (i. e. , detection, semantic/instance segmentation) in autonomous driving domain. This dataset has been used to hold the ICCV2021 SSLAD challenge. More information can refer to https: //soda-2d. github. io.

PDF Details

ICML Conference 2021 Conference Paper

SparseBERT: Rethinking the Importance Analysis in Self-attention

  • Han Shi
  • Jiahui Gao
  • Xiaozhe Ren
  • Hang Xu 0004
  • Xiaodan Liang
  • Zhenguo Li
  • James T. Kwok

Transformer-based models are popularly used in natural language processing (NLP). Its core component, self-attention, has aroused widespread interest. To understand the self-attention mechanism, a direct method is to visualize the attention map of a pre-trained model. Based on the patterns observed, a series of efficient Transformers with different sparse attention masks have been proposed. From a theoretical perspective, universal approximability of Transformer-based models is also recently proved. However, the above understanding and analysis of self-attention is based on a pre-trained model. To rethink the importance analysis in self-attention, we study the significance of different positions in attention matrix during pre-training. A surprising result is that diagonal elements in the attention map are the least important compared with other attention positions. We provide a proof showing that these diagonal elements can indeed be removed without deteriorating model performance. Furthermore, we propose a Differentiable Attention Mask (DAM) algorithm, which further guides the design of the SparseBERT. Extensive experiments verify our interesting findings and illustrate the effect of the proposed algorithm.

Details

NeurIPS Conference 2021 Conference Paper

Towards a Theoretical Framework of Out-of-Distribution Generalization

  • Haotian Ye
  • Chuanlong Xie
  • Tianle Cai
  • Ruichen Li
  • Zhenguo Li
  • Liwei Wang

Generalization to out-of-distribution (OOD) data is one of the central problems in modern machine learning. Recently, there is a surge of attempts to propose algorithms that mainly build upon the idea of extracting invariant features. Although intuitively reasonable, theoretical understanding of what kind of invariance can guarantee OOD generalization is still limited, and generalization to arbitrary out-of-distribution is clearly impossible. In this work, we take the first step towards rigorous and quantitative definitions of 1) what is OOD; and 2) what does it mean by saying an OOD problem is learnable. We also introduce a new concept of expansion function, which characterizes to what extent the variance is amplified in the test domains over the training domains, and therefore give a quantitative meaning of invariant features. Based on these, we prove an OOD generalization error bound. It turns out that OOD generalization largely depends on the expansion function. As recently pointed out by Gulrajani & Lopez-Paz (2020), any OOD learning algorithm without a model selection module is incomplete. Our theory naturally induces a model selection criterion. Extensive experiments on benchmark OOD datasets demonstrate that our model selection criterion has a significant advantage over baselines.

PDF Details

NeurIPS Conference 2020 Conference Paper

Bridging the Gap between Sample-based and One-shot Neural Architecture Search with BONAS

  • Han Shi
  • Renjie Pi
  • Hang Xu
  • Zhenguo Li
  • James Kwok
  • Tong Zhang

Neural Architecture Search (NAS) has shown great potentials in finding better neural network designs. Sample-based NAS is the most reliable approach which aims at exploring the search space and evaluating the most promising architectures. However, it is computationally very costly. As a remedy, the one-shot approach has emerged as a popular technique for accelerating NAS using weight-sharing. However, due to the weight-sharing of vastly different networks, the one-shot approach is less reliable than the sample-based approach. In this work, we propose BONAS (Bayesian Optimized Neural Architecture Search), a sample-based NAS framework which is accelerated using weight-sharing to evaluate multiple related architectures simultaneously. Specifically, we apply Graph Convolutional Network predictor as a surrogate model for Bayesian Optimization to select multiple related candidate models in each iteration. We then apply weight-sharing to train multiple candidate models simultaneously. This approach not only accelerates the traditional sample-based approach significantly, but also keeps its reliability. This is because weight-sharing among related architectures are more reliable than those in the one-shot approach. Extensive experiments are conducted to verify the effectiveness of our method over many competing algorithms.

PDF Details

IJCAI Conference 2020 Conference Paper

DropNAS: Grouped Operation Dropout for Differentiable Architecture Search

  • Weijun Hong
  • Guilin Li
  • Weinan Zhang
  • Ruiming Tang
  • Yunhe Wang
  • Zhenguo Li
  • Yong Yu

Neural architecture search (NAS) has shown encouraging results in automating the architecture design. Recently, DARTS relaxes the search process with a differentiable formulation that leverages weight-sharing and SGD for cost reduction of NAS. In DARTS, all candidate operations are trained simultaneously during the network weight training step. Our empirical results show that this training procedure leads to the co-adaption problem and Matthew Effect: operations with fewer parameters would be trained maturely earlier. This causes two problems: firstly, the operations with more parameters may never have the chance to express the desired function since those with less have already done the job; secondly, the system will punish those underperforming operations by lowering their architecture parameter and backward smaller loss gradients, this causes the Matthew Effect. In this paper, we systematically study these problems and propose a novel grouped operation dropout algorithm named DropNAS to fix the problems with DARTS. Extensive experiments demonstrate that DropNAS solves the above issues and achieves promising performance. Specifically, DropNAS achieves 2. 26% test error on CIFAR-10, 16. 39% on CIFAR-100 and 23. 4% on ImageNet (with the same training hyperparameters as DARTS for a fair comparison). It is also observed that DropNAS is robust across variants of the DARTS search space. Code is available at https: //github. com/huawei-noah.

PDF Details DOI

AAAI Conference 2020 Conference Paper

EHSOD: CAM-Guided End-to-End Hybrid-Supervised Object Detection with Cascade Refinement

  • Linpu Fang
  • Hang Xu
  • Zhili LIU
  • Sarah Parisot
  • Zhenguo Li

Object detectors trained on fully-annotated data currently yield state of the art performance but require expensive manual annotations. On the other hand, weakly-supervised detectors have much lower performance and cannot be used reliably in a realistic setting. In this paper, we study the hybrid-supervised object detection problem, aiming to train a high quality detector with only a limited amount of fullyannotated data and fully exploiting cheap data with imagelevel labels. State of the art methods typically propose an iterative approach, alternating between generating pseudo-labels and updating a detector. This paradigm requires careful manual hyper-parameter tuning for mining good pseudo labels at each round and is quite time-consuming. To address these issues, we present EHSOD, an end-to-end hybrid-supervised object detection system which can be trained in one shot on both fully and weakly-annotated data. Specifically, based on a two-stage detector, we proposed two modules to fully utilize the information from both kinds of labels: 1) CAM- RPN module aims at finding foreground proposals guided by a class activation heat-map; 2) hybrid-supervised cascade module further refines the bounding-box position and classi- fication with the help of an auxiliary head compatible with image-level data. Extensive experiments demonstrate the effectiveness of the proposed method and it achieves comparable results on multiple object detection benchmarks with only 30% fully-annotated data, e. g. 37. 5% mAP on COCO. We will release the code and the trained models.

PDF Details

NeurIPS Conference 2020 Conference Paper

Locally Differentially Private (Contextual) Bandits Learning

  • Kai Zheng
  • Tianle Cai
  • Weiran Huang
  • Zhenguo Li
  • Liwei Wang

We study locally differentially private (LDP) bandits learning in this paper. First, we propose simple black-box reduction frameworks that can solve a large family of context-free bandits learning problems with LDP guarantee. Based on our frameworks, we can improve previous best results for private bandits learning with one-point feedback, such as private Bandits Convex Optimization etc, and obtain the first results for Bandits Convex Optimization (BCO) with multi-point feedback under LDP. LDP guarantee and black-box nature make our frameworks more attractive in real applications compared with previous specifically designed and relatively weaker differentially private (DP) algorithms. Further, we also extend our algorithm to Generalized Linear Bandits with regret bound $\tilde{\mc{O}}(T^{3/4}/\varepsilon)$ under $(\varepsilon, \delta)$-LDP and it is conjectured to be optimal. Note given existing $\Omega(T)$ lower bound for DP contextual linear bandits (Shariff & Sheffet, NeurIPS 2018), our result shows a fundamental difference between LDP and DP for contextual bandits.

PDF Details

AAAI Conference 2020 Conference Paper

Meta-Learning PAC-Bayes Priors in Model Averaging

  • Yimin Huang
  • Weiran Huang
  • Liang Li
  • Zhenguo Li

Nowadays model uncertainty has become one of the most important problems in both academia and industry. In this paper, we mainly consider the scenario in which we have a common model set used for model averaging instead of selecting a single final model via a model selection procedure to account for this model’s uncertainty in order to improve reliability and accuracy of inferences. Here one main challenge is to learn the prior over the model set. To tackle this problem, we propose two data-based algorithms to get proper priors for model averaging. One is for meta-learner, the analysts should use historical similar tasks to extract the information about the prior. The other one is for base-learner, a subsampling method is used to deal with the data step by step. Theoretically, an upper bound of risk for our algorithm is presented to guarantee the performance of the worst situation. In practice, both methods perform well in simulations and real data studies, especially with poor quality data.

PDF Details

AAAI Conference 2020 Conference Paper

New Interpretations of Normalization Methods in Deep Learning

  • Jiacheng Sun
  • Xiangyong Cao
  • Hanwen Liang
  • Weiran Huang
  • Zewei Chen
  • Zhenguo Li

In recent years, a variety of normalization methods have been proposed to help training neural networks, such as batch normalization (BN), layer normalization (LN), weight normalization (WN), group normalization (GN), etc. However, some necessary tools to analyze all these normalization methods are lacking. In this paper, we first propose a lemma to define some necessary tools. Then, we use these tools to make a deep analysis on popular normalization methods and obtain the following conclusions: 1) Most of the normalization methods can be interpreted in a unified framework, namely normalizing pre-activations or weights onto a sphere; 2) Since most of the existing normalization methods are scaling invariant, we can conduct optimization on a sphere with scaling symmetry removed, which can help to stabilize the training of network; 3) We prove that training with these normalization methods can make the norm of weights increase, which could cause adversarial vulnerability as it amplifies the attack. Finally, a series of experiments are conducted to verify these claims.

PDF Details

AAAI Conference 2020 Conference Paper

SM-NAS: Structural-to-Modular Neural Architecture Search for Object Detection

  • Lewei Yao
  • Hang Xu
  • Wei Zhang
  • Xiaodan Liang
  • Zhenguo Li

The state-of-the-art object detection method is complicated with various modules such as backbone, RPN, feature fusion neck and RCNN head, where each module may have different designs and structures. How to leverage the computational cost and accuracy trade-off for the structural combination as well as the modular selection of multiple modules? Neural architecture search (NAS) has shown great potential in finding an optimal solution. Existing NAS works for object detection only focus on searching better design of a single module such as backbone or feature fusion neck, while neglecting the balance of the whole system. In this paper, we present a twostage coarse-to-fine searching strategy named Structural-to- Modular NAS (SM-NAS) for searching a GPU-friendly design of both an efficient combination of modules and better modular-level architecture for object detection. Specifically, Structural-level searching stage first aims to find an efficient combination of different modules; Modular-level searching stage then evolves each specific module and pushes the Pareto front forward to a faster task-specific network. We consider a multi-objective search where the search space covers many popular designs of detection methods. We directly search a detection backbone without pre-trained models or any proxy task by exploring a fast training from scratch strategy. The resulting architectures dominate state-of-the-art object detection systems in both inference time and accuracy and demonstrate the effectiveness on multiple detection datasets, e. g. halving the inference time with additional 1% mAP improvement compared to FPN and reaching 46% mAP with the similar inference time of MaskRCNN.

PDF Details

AAAI Conference 2020 Conference Paper

Universal-RCNN: Universal Object Detector via Transferable Graph R-CNN

  • Hang Xu
  • Linpu Fang
  • Xiaodan Liang
  • Wenxiong Kang
  • Zhenguo Li

The dominant object detection approaches treat each dataset separately and fit towards a specific domain, which cannot adapt to other domains without extensive retraining. In this paper, we address the problem of designing a universal object detection model that exploits diverse category granularity from multiple domains and predict all kinds of categories in one system. Existing works treat this problem by integrating multiple detection branches upon one shared backbone network. However, this paradigm overlooks the crucial semantic correlations between multiple domains, such as categories hierarchy, visual similarity, and linguistic relationship. To address these drawbacks, we present a novel universal object detector called Universal-RCNN that incorporates graph transfer learning for propagating relevant semantic information across multiple datasets to reach semantic coherency. Specifically, we first generate a global semantic pool by integrating all high-level semantic representation of all the categories. Then an Intra-Domain Reasoning Module learns and propagates the sparse graph representation within one dataset guided by a spatial-aware GCN. Finally, an Inter- Domain Transfer Module is proposed to exploit diverse transfer dependencies across all domains and enhance the regional feature representation by attending and transferring semantic contexts globally. Extensive experiments demonstrate that the proposed method significantly outperforms multiple-branch models and achieves the state-of-the-art results on multiple object detection benchmarks (mAP: 49. 1% on COCO).

PDF Details

IJCAI Conference 2019 Conference Paper

Meta Reinforcement Learning with Task Embedding and Shared Policy

  • Lin Lan
  • Zhenguo Li
  • Xiaohong Guan
  • Pinghui Wang

Despite significant progress, deep reinforcement learning (RL) suffers from data-inefficiency and limited generalization. Recent efforts apply meta-learning to learn a meta-learner from a set of RL tasks such that a novel but related task could be solved quickly. Though specific in some ways, different tasks in meta-RL are generally similar at a high level. However, most meta-RL methods do not explicitly and adequately model the specific and shared information among different tasks, which limits their ability to learn training tasks and to generalize to novel tasks. In this paper, we propose to capture the shared information on the one hand and meta-learn how to quickly abstract the specific information about a task on the other hand. Methodologically, we train an SGD meta-learner to quickly optimize a task encoder for each task, which generates a task embedding based on past experience. Meanwhile, we learn a policy which is shared across all tasks and conditioned on task embeddings. Empirical results on four simulated tasks demonstrate that our method has better learning capacity on both training and novel tasks and attains up to 3 to 4 times higher returns compared to baselines.

PDF Details

IJCAI Conference 2017 Conference Paper

DeepFM: A Factorization-Machine based Neural Network for CTR Prediction

  • Huifeng Guo
  • Ruiming Tang
  • Yunming Ye
  • Zhenguo Li
  • Xiuqiang He

Learning sophisticated feature interactions behind user behaviors is critical in maximizing CTR for recommender systems. Despite great progress, existing methods seem to have a strong bias towards low- or high-order interactions, or require expertise feature engineering. In this paper, we show that it is possible to derive an end-to-end learning model that emphasizes both low- and high-order feature interactions. The proposed model, DeepFM, combines the power of factorization machines for recommendation and deep learning for feature learning in a new neural network architecture. Compared to the latest Wide & Deep model from Google, DeepFM has a shared input to its "wide" and "deep" parts, with no need of feature engineering besides raw features. Comprehensive experiments are conducted to demonstrate the effectiveness and efficiency of DeepFM over the existing models for CTR prediction, on both benchmark data and commercial data.

PDF Details

NeurIPS Conference 2013 Conference Paper

Analyzing the Harmonic Structure in Graph-Based Learning

  • Xiao-ming Wu
  • Zhenguo Li
  • Shih-Fu Chang

We show that either explicitly or implicitly, various well-known graph-based models exhibit a common significant \emph{harmonic} structure in its target function -- the value of a vertex is approximately the weighted average of the values of its adjacent neighbors. Understanding of such structure and analysis of the loss defined over such structure help reveal important properties of the target function over a graph. In this paper, we show that the variation of the target function across a cut can be upper and lower bounded by the ratio of its harmonic loss and the cut cost. We use this to develop an analytical tool and analyze 5 popular models in graph-based learning: absorbing random walks, partially absorbing random walks, hitting times, pseudo-inverse of graph Laplacian, and eigenvectors of the Laplacian matrices. Our analysis well explains several open questions of these models reported in the literature. Furthermore, it provides theoretical justifications and guidelines for their practical use. Simulations on synthetic and real datasets support our analysis.

PDF Details

NeurIPS Conference 2012 Conference Paper

Learning with Partially Absorbing Random Walks

  • Xiao-ming Wu
  • Zhenguo Li
  • Anthony So
  • John Wright
  • Shih-Fu Chang

We propose a novel stochastic process that is with probability $\alpha_i$ being absorbed at current state $i$, and with probability $1-\alpha_i$ follows a random edge out of it. We analyze its properties and show its potential for exploring graph structures. We prove that under proper absorption rates, a random walk starting from a set $\mathcal{S}$ of low conductance will be mostly absorbed in $\mathcal{S}$. Moreover, the absorption probabilities vary slowly inside $\mathcal{S}$, while dropping sharply outside $\mathcal{S}$, thus implementing the desirable cluster assumption for graph-based learning. Remarkably, the partially absorbing process unifies many popular models arising in a variety of contexts, provides new insights into them, and makes it possible for transferring findings from one paradigm to another. Simulation results demonstrate its promising applications in graph-based learning.

PDF Details

NeurIPS Conference 2009 Conference Paper

Fast Graph Laplacian Regularized Kernel Learning via Semidefinite–Quadratic–Linear Programming

  • Xiao-ming Wu
  • Anthony So
  • Zhenguo Li
  • Shuo-yen Li

Kernel learning is a powerful framework for nonlinear data modeling. Using the kernel trick, a number of problems have been formulated as semidefinite programs (SDPs). These include Maximum Variance Unfolding (MVU) (Weinberger et al. , 2004) in nonlinear dimensionality reduction, and Pairwise Constraint Propagation (PCP) (Li et al. , 2008) in constrained clustering. Although in theory SDPs can be efficiently solved, the high computational complexity incurred in numerically processing the huge linear matrix inequality constraints has rendered the SDP approach unscalable. In this paper, we show that a large class of kernel learning problems can be reformulated as semidefinite-quadratic-linear programs (SQLPs), which only contain a simple positive semidefinite constraint, a second-order cone constraint and a number of linear constraints. These constraints are much easier to process numerically, and the gain in speedup over previous approaches is at least of the order $m^{2. 5}$, where m is the matrix dimension. Experimental results are also presented to show the superb computational efficiency of our approach.

PDF Details

ICML Conference 2008 Conference Paper

Pairwise constraint propagation by semidefinite programming for semi-supervised classification

  • Zhenguo Li
  • Jianzhuang Liu
  • Xiaoou Tang

We consider the general problem of learning from both pairwise constraints and unlabeled data. The pairwise constraints specify whether two objects belong to the same class or not, known as the must-link constraints and the cannot-link constraints. We propose to learn a mapping that is smooth over the data graph and maps the data onto a unit hypersphere, where two must-link objects are mapped to the same point while two cannot-link objects are mapped to be orthogonal. We show that such a mapping can be achieved by formulating a semidefinite programming problem, which is convex and can be solved globally. Our approach can effectively propagate pairwise constraints to the whole data set. It can be directly applied to multi-class classification and can handle data labels, pairwise constraints, or a mixture of them in a unified framework. Promising experimental results are presented for classification tasks on a variety of synthetic and real data sets.

Details
v2026.09.13