Dongsheng Li

EAAI Journal 2026 Journal Article

An automated framework for converting point cloud data to building information modeling with segmentation and refinement

• Tianze Chen
• Hongxu Wang
• Dongsheng Li
• Jiepeng Liu
• Pengkun Liu
• Zhou Wu
• Chengran Xu
• Meifei Zhang

AAAI Conference 2026 Conference Paper

Improving Long-Context Summarization with Multi-Granularity Retrieval Optimization

• Xueyu Chen
• Kaitao Song
• Zifan Song
• Dongsheng Li
• Cairong Zhao

Retrieval-Augmented Generation (RAG) is an effective solution to overcome the limitations of Large Language Models (LLMs) in terms of specific-domain knowledge and timely information updates. However, current RAG methods typically respond to queries based on isolated segments, lacking the ability to integrate information within the same document. This undermines performance in real-world tasks requiring coherent understanding across an entire document. Notably, the human brain naturally integrates and summarizes prior knowledge upon reading a given text, progressively formulating a comprehensive understanding. Motivated by this cognitive process, we propose the Hierarchical Two-Stage Summarization-based Information Retrieval (HTSIR) method, which preprocesses the corpus prior to retrieval, summarizes continuous texts to obtain integrated information, and constructs a retrieval tree with varying summary granularities. The retrieved information is then processed by a Reranker based on the current question to serve as a context for LLMs. Additionally, as single-step summarization is often imprecise in query-based summarization tasks, we further apply a Refinement module, allowing LLMs to reflect and revise their output to achieve the final result. By combining HTSIR with GPT-4o mini, we achieve state-of-the-art results on complex question tasks across four long-text datasets (NarrativeQA, QASPER, QuALITY, and QMSum), achieving an improvement of about 6 points on the Question Answering (QA) task in QuALITY-HRAD.

AAAI Conference 2026 Conference Paper

WALKSAFE: Risk-aware Graph Random Walk with Bi-GRPO for LLM Safety

• Shilong Pan
• Zhiliang Tian
• Wanlong Yu
• Zhen Huang
• Qingyu Qiu
• Zihan Chen
• Zhonghao Sun
• Minlie Huang

Large language models (LLMs) may generate harmful outputs on malicious inputs. Existing safety methods, including prompt engineering and model editing, rely on hand-crafted templates or target-driven parameter modifications, limiting their generalizability in unseen harmful scenarios. Post-training aims to ensure LLM safety in general domains via supervised fine-tuning (SFT) or reinforcement learning (RL) on diverse malicious inputs. SFT needs annotated refusal samples while RL learns to refuse risk by exploring diverse harmful inputs. However, these methods tend to harshly refuse over any possible risks, sacrificing potentially useful information and degrading model utility. We argue that realistic malicious inputs often mix both harmful and helpful semantics (i.e., entities and relations), and LLMs should identify and remove only harmful relations while preserving useful ones. Thus, the original malicious user inputs can shift into safe queries, to which LLMs can respond safely and helpfully. In this paper, we propose WALKSAFE, a graph-based risk-aware training framework that enables LLMs to identify potential risks of key semantics (entities and relations) in user inputs via graph structure. By filtering harmful relations, LLMs can respond to safe input queries and then generate their corresponding safe and helpful responses. First, we model all entities and relations in the inputs with a graph structure. Second, we adopt a risk-aware random walk on the graph to quantify potential risk under multiple entities and relations. Then, we reconstruct safe queries by filtering harmful relations to promote the LLM to answer safely and helpfully rather than with direct refusals. Finally, we propose Bi-GRPO to post-train LLMs. As vanilla GRPO conducts only the intra-group comparison, Bi-GRPO performs both intra-group and inter-group comparisons between different response groups. The extra inter-group rewards encourage the model to distinguish harmful and safe semantics, and thus prefer safe and helpful responses. Experiments on three LLMs show that our models obtain SOTA results.

NeurIPS Conference 2025 Conference Paper

Accelerating Block Coordinate Descent for LLM Finetuning via Landscape Expansion

• Qijun Luo
• Yifei Shen
• Liangzu Peng
• Dongsheng Li
• Xiao Li

Finetuning large language models (LLMs) is a resource-intensive task for researchers in academia, with memory constraints posing a key bottleneck. A classic optimization method, block coordinate descent (BCD), significantly reduces memory cost by segmenting the trainable parameters into multiple blocks and optimizing one active block at a time while freezing the others. However, we identify that blindly applying BCD to train LLMs can be inefficient for two reasons. First, optimizing only the active block requires backpropagating through multiple deeper yet inactive blocks, resulting in wasteful computations. Second, the frozen blocks, when they are not quite close to optimality, can narrow the optimization landscape, potentially misguiding the training of the active block. To address these issues simultaneously, we propose integrating BCD with landscape expansion, which unfreezes the inactive blocks and updates them in a cost-efficient manner during the same backpropagation as the update to the active block. Experiments on 8B and 70B models demonstrate that our proposed method surpasses memory-efficient baselines and matches Adam's downstream performance while requiring only 24 GB of memory for the 8B model and 300 GB for the 70B model.

NeurIPS Conference 2025 Conference Paper

Chain-of-Model Learning for Language Model

• Xiaohua Wang
• Kaitao Song
• Xu Tan
• Huiqiang Jiang
• Chengruidong Zhang
• Yongliang Shen
• Cen Lu
• Zihao Li

In this paper, we propose a novel learning paradigm, termed Chain-of-Model (CoM), which incorporates the causal relationship into the hidden states of each layer as a chain style. thereby introducing great scaling efficiency in model training and inference flexibility in deployment. We introduce the concept of Chain-of-Representation (CoR), which formulates the hidden states at each layer as a combination of multiple sub-representations (i. e. , chains). In each layer, each chain from the output representations can only view all of its preceding chains in the input representations. Consequently, the model built upon CoM framework can progressively scale up the model size by increasing the chains based on the previous models (i. e. , chains), and offer multiple sub-models at varying sizes for elastic inference by using different chain numbers. Based on this principle, we devise Chain-of-Language-Model (CoLM), which incorporates the idea of CoM into each layer of Transformer architecture. Based on CoLM, we further introduce CoLM-Air by introducing a KV sharing mechanism, that computes all keys and values within the first chain and then shares across all chains. This design demonstrates additional extensibility, such as enabling seamless LM switching, prefilling acceleration and so on. Experimental results demonstrate our CoLM family can achieve comparable performance to the standard Transformer, while simultaneously enabling greater flexiblity, such as progressive scaling to improve training efficiency and offer multiple varying model sizes for elastic inference, paving a a new way toward building language models.

TMLR Journal 2025 Journal Article

Graph Theory-Based Deep Graph Similarity Learning: A Unified Survey of Pipeline, Techniques, and Challenges

• Zhouyang LIU
• Ning Liu
• Yixin Chen
• Ziqing Wen
• Jiezhong He
• Dongsheng Li

Graph similarity computation, which measures the resemblance between graphs, is a crucial operation in fields such as graph search. Recent advances in graph neural networks have enabled the embedding of graphs into low-dimensional vector spaces, where the sim- ilarity or distance between graphs can be efficiently quantified. However, these methods are often tailored to specific tasks and function as black boxes, limiting both generalization and interpretability. To address these challenges, there is growing interest in incorporating domain-agnostic and interpretable concepts from graph theory—such as subgraph isomorphism, maximum common subgraph, and graph edit distance—into graph similarity learning as training objectives. This survey presents a comprehensive review of recent advancements in deep graph similarity learning, focusing on models that integrate these graph theory concepts. Despite the different training objectives of these approaches, they share significant commonalities in the training pipeline, techniques, and challenges. We analyze them within a unified lens referred to as graph theory-based deep similarity learning (GTDGSL) methods. We systematically compare existing GTDGSL methods alongside their common training pipeline, highlighting the technique trend and discussing key challenges, applications, and future research directions in this domain. We organize the papers included in this survey and their open-source implementations at https://github.com/liuzhouyang/Graph-Theory-Based-Deep-Graph-Similarity-Learning-Survey.

NeurIPS Conference 2025 Conference Paper

Hippocampal-like Sequential Editing for Continual Knowledge Updates in Large Language Models

• Quntian Fang
• Zhen Huang
• Zhiliang Tian
• Minghao Hu
• Dongsheng Li
• Yiping Yao
• Xinyue Fang
• Menglong Lu

Large language models (LLMs) are now pivotal in real-world applications. Model editing has emerged as a promising paradigm for efficiently modifying LLMs without full retraining. However, current editing approaches face significant limitations due to parameter drift, which stems from inconsistencies between newly edited knowledge and the model's existing knowledge. In sequential editing scenarios, cumulative drifts progressively lead to model collapse characterized by general capability degradation and balance between acquiring new knowledge and catastrophic forgetting of existing knowledge. Drawing inspiration from the hippocampal trisynaptic circuit for continual memorizing and forgetting, we propose a Hippocampal-like Sequential Editing (HSE) framework that designs the unlearning of obsolete knowledge, domain-specific knowledge update separation and replay for edited knowledge. Specifically, the HSE framework designs three core mechanisms: (1) Machine unlearning selectively erases outdated knowledge to facilitate integration of new information, (2) Fisher Information Matrix-guided parameter updates prevents cross-domain knowledge interference, and (3) Parameter replay consolidates long-term editing memory through lightweight and global replay of editing data in a parametric form. Theoretical analysis demonstrates that HSE achieves smaller generalization error bounds, more stable convergence and higher computational efficiency. Experimental results validate its effective balance between acquiring new knowledge and mitigating catastrophic forgetting, maintaining or even slightly enhancing general capabilities. In practical applications, experiments confirm its effectiveness in multi-domain hallucination mitigation, healthcare knowledge injecting, and societal bias reduction.

NeurIPS Conference 2025 Conference Paper

Learning to Instruct for Visual Instruction Tuning

• Zhihan Zhou
• Feng Hong
• JIAAN LUO
• Yushi Ye
• Jiangchao Yao
• Dongsheng Li
• Bo Han
• Ya Zhang

We propose L2T, an advancement of visual instruction tuning (VIT). While VIT equips Multimodal LLMs (MLLMs) with promising multimodal capabilities, the current design choices for VIT often result in overfitting and shortcut learning, potentially degrading performance. This gap arises from an overemphasis on instruction-following abilities, while neglecting the proactive understanding of visual information. Inspired by this, L2T adopts a simple yet effective approach by incorporating the loss function into both the instruction and response sequences. It seamlessly expands the training data, and regularizes the MLLMs from overly relying on language priors. Based on this merit, L2T achieves a significant relative improvement of up to 9% on comprehensive multimodal benchmarks, requiring no additional training data and incurring negligible computational overhead. Surprisingly, L2T attains exceptional fundamental visual capabilities, yielding up to an 18% improvement in captioning performance, while simultaneously alleviating hallucination in MLLMs. Github code: https: //github. com/Feng-Hong/L2T.

NeurIPS Conference 2025 Conference Paper

Omni-DNA: A Genomic Model Supporting Sequence Understanding, Long-context, and Textual Annotation

• Zehui Li
• Vallijah Subasri
• Yifei Shen
• Dongsheng Li
• Wentao Gu
• Guy-Bart Stan
• Yiren Zhao
• Caihua Shan

The interpretation of genomic sequences is crucial for understanding biological processes. To handle the growing volume of DNA sequence data, Genomic Foundation Models (GFMs) have been developed by adapting architectures and training paradigms from Large Language Models (LLMs). Despite their remarkable performance in DNA sequence classification tasks, there remains a lack of systematic understanding regarding the training and task-adaptation processes of GFMs. Moreover, existing GFMs cannot achieve state-of-the-art performance on both short and long-context tasks and lacks multimodal abilities. By revisiting pre-training architectures and post-training techniques, we propose Omni-DNA, a family of models spanning 20M to 1. 1B parameters that supports sequence understanding, long-context genomic reasoning, and natural-language annotation. Omni-DNA establishes new state-of-the-art results on 18 of 26 evaluations drawn from Nucleotide Transformer and Genomic Benchmarks. When jointly fine-tuning on biologically related tasks, Omni-DNA consistently outperform existing models and demonstrate multi-tasking abilities. To enable processing of arbitrary sequence lengths, we introduce SEQPACK —an adaptive compression operator that packs historical tokens into a learned synopsis using a position-aware learnable sampling mechanism, enabling transformer-based models to process ultra-long sequences with minimal memory and computational requirements. Our approach demonstrates superior performance on enhancer-target interaction tasks, capturing distant regulatory interactions at the 450kbp range more effectively than existing models. Finally, we present a new dataset termed seq2func, enabling Omni-DNA to generate accurate and functionally meaningful interpretations of DNA sequences, unlocking new possibilities for genomic analysis and discovery.

NeurIPS Conference 2025 Conference Paper

SimSort: A Data-Driven Framework for Spike Sorting by Large-Scale Electrophysiology Simulation

• Yimu Zhang
• Dongqi Han
• Yansen Wang
• Zhenning Lv
• Yu Gu
• Dongsheng Li

Spike sorting is an essential process in neural recording, which identifies and separates electrical signals from individual neurons recorded by electrodes in the brain, enabling researchers to study how specific neurons communicate and process information. Although there exist a number of spike sorting methods which have contributed to significant neuroscientific breakthroughs, many are heuristically designed, making it challenging to verify their correctness due to the difficulty of obtaining ground truth labels from real-world neural recordings. In this work, we explore a data-driven, deep learning-based approach. We begin by creating a large-scale dataset through electrophysiology simulations using biologically realistic computational models. We then present SimSort, a pretraining framework for spike sorting. Trained solely on simulated data, SimSort demonstrates zero-shot generalizability to real-world spike sorting tasks, yielding consistent improvements over existing methods across multiple benchmarks. These results highlight the potential of simulation-driven pretraining to enhance the robustness and scalability of spike sorting in experimental neuroscience.

NeurIPS Conference 2025 Conference Paper

Toward Relative Positional Encoding in Spiking Transformers

• Changze Lv
• Yansen Wang
• Dongqi Han
• Yifei Shen
• Xiaoqing Zheng
• Xuanjing Huang
• Dongsheng Li

Spiking neural networks (SNNs) are bio-inspired networks that mimic how neurons in the brain communicate through discrete spikes, which have great potential in various tasks due to their energy efficiency and temporal processing capabilities. SNNs with self-attention mechanisms (spiking Transformers) have recently shown great advancements in various tasks, and inspired by traditional Transformers, several studies have demonstrated that spiking absolute positional encoding can help capture sequential relationships for input data, enhancing the capabilities of spiking Transformers for tasks such as sequential modeling and image classification. However, how to incorporate relative positional information into SNNs remains a challenge. In this paper, we introduce several strategies to approximate relative positional encoding (RPE) in spiking Transformers while preserving the binary nature of spikes. Firstly, we formally prove that encoding relative distances with Gray Code ensures that the binary representations of positional indices maintain a constant Hamming distance whenever their decimal values differ by a power of two, and we propose Gray-PE based on this property. In addition, we propose another RPE method called Log-PE, which combines the logarithmic form of the relative distance matrix directly into the spiking attention map. Furthermore, we extend our RPE methods to a two-dimensional form, making them suitable for processing image patches. We evaluate our RPE methods on various tasks, including time series forecasting, text classification, and patch-based image classification, and the experimental results demonstrate a satisfying performance gain by incorporating our RPE methods across many architectures. Our results provide fresh perspectives on designing spiking Transformers to advance their sequential modeling capability, thereby expanding their applicability across various domains. Our code is available at https: //github. com/microsoft/SeqSNN.

AAAI Conference 2025 Conference Paper

Zero-resource Hallucination Detection for Text Generation via Graph-based Contextual Knowledge Triples Modeling

• Xinyue Fang
• Zhen Huang
• Zhiliang Tian
• Minghui Fang
• Ziyi Pan
• Quntian Fang
• Zhihua Wen
• Hengyue Pan

LLMs obtain remarkable performance but suffer from hallucinations. Most research on detecting hallucination focuses on questions with short and concrete correct answers that are easy to check faithfulness. Hallucination detections for text generation with open-ended answers are more hard. Some researchers use external knowledge to detect hallucinations in generated texts, but external resources for specific scenarios are hard to access. Recent studies on detecting hallucinations in long texts without external resources conduct consistency comparison among multiple sampled outputs. To handle long texts, researchers split long texts into multiple facts and individually compare the consistency of each pair of facts. However, these methods (1) hardly achieve alignment among multiple facts; (2) overlook dependencies between multiple contextual facts. In this paper, we propose a graph-based context-aware (GCA) hallucination detection method for text generations, which aligns facts and considers the dependencies between contextual facts in consistency comparison. Particularly, to align multiple facts, we conduct a triple-oriented response segmentation to extract multiple knowledge triples. To model dependencies among contextual triples (facts), we construct contextual triples into a graph and enhance triples’ interactions via message passing and aggregating via RGCN. To avoid the omission of knowledge triples in long texts, we conduct an LLM-based reverse verification by reconstructing the knowledge triples. Experiments show that our model enhances hallucination detection and excels all baselines.

NeurIPS Conference 2024 Conference Paper

Advancing Spiking Neural Networks for Sequential Modeling with Central Pattern Generators

• Changze Lv
• Dongqi Han
• Yansen Wang
• Xiaoqing Zheng
• Xuanjing Huang
• Dongsheng Li

Spiking neural networks (SNNs) represent a promising approach to developing artificial neural networks that are both energy-efficient and biologically plausible. However, applying SNNs to sequential tasks, such as text classification and time-series forecasting, has been hindered by the challenge of creating an effective and hardware-friendly spike-form positional encoding (PE) strategy. Drawing inspiration from the central pattern generators (CPGs) in the human brain, which produce rhythmic patterned outputs without requiring rhythmic inputs, we propose a novel PE technique for SNNs, termed CPG-PE. We demonstrate that the commonly used sinusoidal PE is mathematically a specific solution to the membrane potential dynamics of a particular CPG. Moreover, extensive experiments across various domains, including time-series forecasting, natural language processing, and image classification, show that SNNs with CPG-PE outperform their conventional counterparts. Additionally, we perform analysis experiments to elucidate the mechanism through which SNNs encode positional information and to explore the function of CPGs in the human brain. This investigation may offer valuable insights into the fundamental principles of neural computation.

AAAI Conference 2024 Conference Paper

Auto-Prox: Training-Free Vision Transformer Architecture Search via Automatic Proxy Discovery

• Zimian Wei
• Peijie Dong
• Zheng Hui
• Anggeng Li
• Lujun Li
• Menglong Lu
• Hengyue Pan
• Dongsheng Li

The substantial success of Vision Transformer (ViT) in computer vision tasks is largely attributed to the architecture design. This underscores the necessity of efficient architecture search for designing better ViTs automatically. As training-based architecture search methods are computationally intensive, there’s a growing interest in training-free methods that use zero-cost proxies to score ViTs. However, existing training-free approaches require expert knowledge to manually design specific zero-cost proxies. Moreover, these zero-cost proxies exhibit limitations to generalize across diverse domains. In this paper, we introduce Auto-Prox, an automatic proxy discovery framework, to address the problem. First, we build the ViT-Bench-101, which involves different ViT candidates and their actual performance on multiple datasets. Utilizing ViT-Bench-101, we can evaluate zero-cost proxies based on their score-accuracy correlation. Then, we represent zero-cost proxies with computation graphs and organize the zero-cost proxy search space with ViT statistics and primitive operations. To discover generic zero-cost proxies, we propose a joint correlation metric to evolve and mutate different zero-cost proxy candidates. We introduce an elitism-preserve strategy for search efficiency to achieve a better trade-off between exploitation and exploration. Based on the discovered zero-cost proxy, we conduct a ViT architecture search in a training-free manner. Extensive experiments demonstrate that our method generalizes well to different datasets and achieves state-of-the-art results both in ranking correlation and final accuracy. Codes can be found at https://github.com/lilujunai/Auto-Prox-AAAI24.

NeurIPS Conference 2024 Conference Paper

Can Graph Learning Improve Planning in LLM-based Agents?

• Xixi Wu
• Yifei Shen
• Caihua Shan
• Kaitao Song
• Siwei Wang
• Bohang Zhang
• Jiarui Feng
• Hong Cheng

Task planning in language agents is emerging as an important research topic alongside the development of large language models (LLMs). It aims to break down complex user requests in natural language into solvable sub-tasks, thereby fulfilling the original requests. In this context, the sub-tasks can be naturally viewed as a graph, where the nodes represent the sub-tasks, and the edges denote the dependencies among them. Consequently, task planning is a decision-making problem that involves selecting a connected path or subgraph within the corresponding graph and invoking it. In this paper, we explore graph learning-based methods for task planning, a direction that is orthogonal to the prevalent focus on prompt design. Our interest in graph learning stems from a theoretical discovery: the biases of attention and auto-regressive loss impede LLMs' ability to effectively navigate decision-making on graphs, which is adeptly addressed by graph neural networks (GNNs). This theoretical insight led us to integrate GNNs with LLMs to enhance overall performance. Extensive experiments demonstrate that GNN-based methods surpass existing solutions even without training, and minimal training can further enhance their performance. The performance gain increases with a larger task graph size.

NeurIPS Conference 2024 Conference Paper

EEG2Video: Towards Decoding Dynamic Visual Perception from EEG Signals

• Xuan-Hao Liu
• Yan-Kai Liu
• Yansen Wang
• Kan Ren
• Hanwen Shi
• Zilong Wang
• Dongsheng Li
• Bao-Liang Lu

Our visual experience in daily life are dominated by dynamic change. Decoding such dynamic information from brain activity can enhance the understanding of the brain’s visual processing system. However, previous studies predominately focus on reconstructing static visual stimuli. In this paper, we explore to decode dynamic visual perception from electroencephalography (EEG), a neuroimaging technique able to record brain activity with high temporal resolution (1000 Hz) for capturing rapid changes in brains. Our contributions are threefold: Firstly, we develop a large dataset recording signals from 20 subjects while they were watching 1400 dynamic video clips of 40 concepts. This dataset fills the gap in the lack of EEG-video pairs. Secondly, we annotate each video clips to investigate the potential for decoding some specific meta information (e. g. , color, dynamic, human or not) from EEG. Thirdly, we propose a novel baseline EEG2Video for video reconstruction from EEG signals that better aligns dynamic movements with high temporal resolution brain signals by Seq2Seq architecture. EEG2Video achieves a 2-way accuracy of 79. 8% in semantic classification tasks and 0. 256 in structural similarity index (SSIM). Overall, our works takes an important step towards decoding dynamic visual perception from EEG signals. Our dataset and code will be released soon.

IJCAI Conference 2024 Conference Paper

Exploring the Inefficiency of Heavy Ball as Momentum Parameter Approaches 1

• Xiaoge Deng
• Tao Sun
• Dongsheng Li
• Xicheng Lu

The heavy ball momentum method is a commonly used technique for accelerating training processes in the machine learning community. However, empirical evidence suggests that the convergence of stochastic gradient descent (SGD) with heavy ball may slow down when the momentum hyperparameter approaches 1. Despite this observation, there are no established theories or solutions to explain and address this issue. In this study, we provide the first theoretical result that elucidates why momentum slows down SGD as it tends to 1. To better understand this inefficiency, we focus on the quadratic convex objective in the analysis. Our findings show that momentum accelerates SGD when the scaling parameter is not very close to 1. Conversely, when the scaling parameter approaches 1, momentum impairs SGD and degrades its stability. Based on the theoretical findings, we propose a descending warmup technique for the heavy ball momentum, which exploits the advantages of the heavy ball method and overcomes the inefficiency problem when the momentum tends to 1. Numerical results demonstrate the effectiveness of the proposed SHB-DW algorithm.

AAAI Conference 2024 Conference Paper

Learning Hierarchical Prompt with Structured Linguistic Knowledge for Vision-Language Models

• Yubin Wang
• Xinyang Jiang
• De Cheng
• Dongsheng Li
• Cairong Zhao

Prompt learning has become a prevalent strategy for adapting vision-language foundation models to downstream tasks. As large language models (LLMs) have emerged, recent studies have explored the use of category-related descriptions as input to enhance prompt effectiveness. Nevertheless, conventional descriptions fall short of structured information that effectively represents the interconnections among entities or attributes linked to a particular category. To address this limitation and prioritize harnessing structured knowledge, this paper advocates for leveraging LLMs to build a graph for each description to model the entities and attributes describing the category, as well as their correlations. Preexisting prompt tuning methods exhibit inadequacies in managing this structured knowledge. Consequently, we propose a novel approach called Hierarchical Prompt Tuning (HPT), which enables simultaneous modeling of both structured and conventional linguistic knowledge. Specifically, we introduce a relationship-guided attention module to capture pair-wise associations among entities and attributes for low-level prompt learning. In addition, by incorporating high-level and global-level prompts modeling overall semantics, the proposed hierarchical structure forges cross-level interlinks and empowers the model to handle more complex and long-term relationships. Extensive experiments demonstrate that our HPT shows strong effectiveness and generalizes much better than existing SOTA methods. Our code is available at https://github.com/Vill-Lab/2024-AAAI-HPT.

IJCAI Conference 2024 Conference Paper

LLMs Can Find Mathematical Reasoning Mistakes by Pedagogical Chain-of-Thought

• Zhuoxuan Jiang
• Haoyuan Peng
• Shanshan Feng
• Fan Li
• Dongsheng Li

Self-correction is emerging as a promising approach to mitigate the issue of hallucination in Large Language Models (LLMs). To facilitate effective self-correction, recent research has proposed mistake detection as its initial step. However, current literature suggests that LLMs often struggle with reliably identifying reasoning mistakes when using simplistic prompting strategies. To address this challenge, we introduce a unique prompting strategy, termed the Pedagogical Chain-of-Thought (PedCoT), which is specifically designed to guide the identification of reasoning mistakes, particularly mathematical reasoning mistakes. PedCoT consists of pedagogical principles for prompts (PPP) design, two-stage interaction process (TIP) and grounded PedCoT prompts, all inspired by the educational theory of the Bloom Cognitive Model (BCM). We evaluate our approach on two public datasets featuring math problems of varying difficulty levels. The experiments demonstrate that our zero-shot prompting strategy significantly outperforms strong baselines. The proposed method can achieve the goal of reliable mathematical mistake identification and provide a foundation for automatic math answer grading. The results underscore the significance of educational theory, serving as domain knowledge, in guiding prompting strategy design for addressing challenging tasks with LLMs effectively.

NeurIPS Conference 2024 Conference Paper

MInference 1.0: Accelerating Pre-filling for Long-Context LLMs via Dynamic Sparse Attention

• Huiqiang Jiang
• Yucheng Li
• Chengruidong Zhang
• Qianhui Wu
• Xufang Luo
• Surin Ahn
• Zhenhua Han
• Amir H. Abdi

The computational challenges of Large Language Model (LLM) inference remain a significant barrier to their widespread deployment, especially as prompt lengths continue to increase. Due to the quadratic complexity of the attention computation, it takes 30 minutes for an 8B LLM to process a prompt of 1M tokens (i. e. , the pre-filling stage) on a single A100 GPU. Existing methods for speeding up prefilling often fail to maintain acceptable accuracy or efficiency when applied to long-context LLMs. To address this gap, we introduce MInference (Milliontokens Inference), a sparse calculation method designed to accelerate pre-filling of long-sequence processing. Specifically, we identify three unique patterns in long-context attention matrices-the A-shape, Vertical-Slash, and Block-Sparse-that can be leveraged for efficient sparse computation on GPUs. We determine the optimal pattern for each attention head offline and dynamically build sparseindices based on the assigned pattern during inference. With the pattern and sparse indices, we perform efficient sparse attention calculations via our optimized GPU kernels to significantly reduce the latency in the pre-filling stage of longcontext LLMs. Our proposed technique can be directly applied to existing LLMs without any modifications to the pre-training setup or additional fine-tuning. Byevaluating on a wide range of downstream tasks, including InfiniteBench, RULER, PG-19, and Needle In A Haystack, and models including LLaMA-3-1M, GLM-4-1M, Yi-200K, Phi-3-128K, and Qwen2-128K, we demonstrate that MInference effectively reduces inference latency by up to 10x for pre-filling on an A100, while maintaining accuracy. Our code is available at https: //aka. ms/MInference.

NeurIPS Conference 2024 Conference Paper

Once Read is Enough: Domain-specific Pretraining-free Language Models with Cluster-guided Sparse Experts for Long-tail Domain Knowledge

• Fang Dong
• Mengyi Chen
• Jixian Zhou
• Yubin Shi
• Yixuan Chen
• Mingzhi Dong
• Yujiang Wang
• Dongsheng Li

Language models (LMs) only pretrained on a general and massive corpus usually cannot attain satisfying performance on domain-specific downstream tasks, and hence, applying domain-specific pretraining to LMs is a common and indispensable practice. However, domain-specific pretraining can be costly and time-consuming, hindering LMs' deployment in real-world applications. In this work, we consider the incapability to memorize domain-specific knowledge embedded in the general corpus with rare occurrences and long-tail distributions as the leading cause for pretrained LMs' inferior downstream performance. Analysis of Neural Tangent Kernels (NTKs) reveals that those long-tail data are commonly overlooked in the model's gradient updates and, consequently, are not effectively memorized, leading to poor domain-specific downstream performance. Based on the intuition that data with similar semantic meaning are closer in the embedding space, we devise a Cluster-guided Sparse Expert (CSE) layer to actively learn long-tail domain knowledge typically neglected in previous pretrained LMs. During pretraining, a CSE layer efficiently clusters domain knowledge together and assigns long-tail knowledge to designate extra experts. CSE is also a lightweight structure that only needs to be incorporated in several deep layers. With our training strategy, we found that during pretraining, data of long-tail knowledge gradually formulate isolated, outlier clusters in an LM's representation spaces, especially in deeper layers. Our experimental results show that only pretraining CSE-based LMs is enough to achieve superior performance than regularly pretrained-finetuned LMs on various downstream tasks, implying the prospects of domain-specific-pretraining-free language models.

NeurIPS Conference 2024 Conference Paper

Perception of Knowledge Boundary for Large Language Models through Semi-open-ended Question Answering

• Zhihua Wen
• Zhiliang Tian
• Zexin Jian
• Zhen Huang
• Pei Ke
• Yifu Gao
• Minlie Huang
• Dongsheng Li

Large Language Models (LLMs) are widely used for knowledge-seeking purposes yet suffer from hallucinations. The knowledge boundary of an LLM limits its factual understanding, beyond which it may begin to hallucinate. Investigating the perception of LLMs' knowledge boundary is crucial for detecting hallucinations and LLMs' reliable generation. Current studies perceive LLMs' knowledge boundary on questions with concrete answers (close-ended questions) while paying limited attention to semi-open-ended questions that correspond to many potential answers. Some researchers achieve it by judging whether the question is answerable or not. However, this paradigm is not so suitable for semi-open-ended questions, which are usually ``partially answerable questions'' containing both answerable answers and ambiguous (unanswerable) answers. Ambiguous answers are essential for knowledge-seeking, but it may go beyond the knowledge boundary of LLMs. In this paper, we perceive the LLMs' knowledge boundary with semi-open-ended questions by discovering more ambiguous answers. First, we apply an LLM-based approach to construct semi-open-ended questions and obtain answers from a target LLM. Unfortunately, the output probabilities of mainstream black-box LLMs are inaccessible to sample more low-probability ambiguous answers. Therefore, we apply an open-sourced auxiliary model to explore ambiguous answers for the target LLM. We calculate the nearest semantic representation for existing answers to estimate their probabilities, with which we reduce the generation probability of high-probability existing answers to achieve a more effective generation. Finally, we compare the results from the RAG-based evaluation and LLM self-evaluation to categorize four types of ambiguous answers that are beyond the knowledge boundary of the target LLM. Following our method, we construct a dataset to perceive the knowledge boundary for GPT-4. We find that GPT-4 performs poorly on semi-open-ended questions and is often unaware of its knowledge boundary. Besides, our auxiliary model, LLaMA-2-13B, is effective in discovering many ambiguous answers, including correct answers neglected by GPT-4 and delusive wrong answers GPT-4 struggles to identify.

YNICL Journal 2024 Journal Article

Relationship of irisin with disease severity and dopamine uptake in Parkinson's disease patients

• Xiaoxue Shi
• Qi Gu
• Chang Fu
• Jianjun Ma
• Dongsheng Li
• Jinhua Zheng
• Siyuan Chen
• Zonghan She

BACKGROUND: This study was designed to investigate the relationship of irisin with the severity of Parkinson's disease (PD) and dopamine (DOPA) uptake in patients with PD and to understand the role of irisin in PD. METHODS: The plasma levels of irisin and α-syn were measured by enzyme-linked immunosorbent assay (ELISA). Motor and nonmotor symptoms were assessed with the relevant scales. DOPA uptake was measured with DOPA positron emission tomography (PET)/magnetic resonance imaging (MRI). RESULTS: The plasma levels of α-syn and irisin in patients with PD gradually increased and decreased, respectively, with the progression of the disease. There was a negative correlation between plasma α-syn and irisin levels in patients with PD. The level of irisin in plasma was negatively correlated with Unified Parkinson's Disease Rating Scale (UPDRS)-III scores and positively correlated with Montreal Cognitive Assessment (MoCA) scores. The striatal/occipital lobe uptake ratios (SORs) of the ipsilateral and contralateral caudate nucleus and anterior and posterior putamen in the high-irisin group were significantly higher than those in the low-irisin group, and irisin levels in the caudate nucleus and anterior and posterior putamen contralateral to the affected limb were lower than those on the ipsilateral side. The level of irisin was positively correlated with the SORs of the ipsilateral and contralateral caudate nucleus and putamen in PD patients. CONCLUSIONS: Irisin plays a neuroprotective role by decreasing the level of α-syn. Irisin is negatively correlated with the severity of motor symptoms and cognitive impairment. More importantly, irisin can improve DOPA uptake in the striatum of patients with PD, especially on the side contralateral to the affected limb.

NeurIPS Conference 2024 Conference Paper

Stability and Generalization of Asynchronous SGD: Sharper Bounds Beyond Lipschitz and Smoothness

• Xiaoge Deng
• Tao Sun
• Shengwei Li
• Dongsheng Li
• Xicheng Lu

Asynchronous stochastic gradient descent (ASGD) has evolved into an indispensable optimization algorithm for training modern large-scale distributed machine learning tasks. Therefore, it is imperative to explore the generalization performance of the ASGD algorithm. However, the existing results are either pessimistic and vacuous or restricted by strict assumptions that fail to reveal the intrinsic impact of asynchronous training on generalization. In this study, we establish sharper stability and generalization bounds for ASGD under much weaker assumptions. Firstly, this paper studies the on-average model stability of ASGD and provides a non-vacuous upper bound on the generalization error, without relying on the Lipschitz assumption. Furthermore, we investigate the excess generalization error of the ASGD algorithm, revealing the effects of asynchronous delay, model initialization, number of training samples and iterations on generalization performance. Secondly, for the first time, this study explores the generalization performance of ASGD in the non-smooth case. We replace smoothness with the much weaker Hölder continuous assumption and achieve similar generalization results as in the smooth case. Finally, we validate our theoretical findings by training numerous machine learning models, including convex problems and non-convex tasks in computer vision and natural language processing.

NeurIPS Conference 2024 Conference Paper

TaskBench: Benchmarking Large Language Models for Task Automation

• Yongliang Shen
• Kaitao Song
• Xu Tan
• Wenqi Zhang
• Kan Ren
• Siyu Yuan
• Weiming Lu
• Dongsheng Li

In recent years, the remarkable progress of large language models (LLMs) has sparked interest in task automation, which involves decomposing complex tasks described by user instructions into sub-tasks and invoking external tools to execute them, playing a central role in autonomous agents. However, there is a lack of systematic and standardized benchmarks to promote the development of LLMs in task automation. To address this, we introduce TaskBench, a comprehensive framework to evaluate the capability of LLMs in task automation. Specifically, task automation can be divided into three critical stages: task decomposition, tool selection, and parameter prediction. To tackle the complexities inherent in these stages, we introduce the concept of Tool Graph to represent decomposed tasks and adopt a back-instruct method to generate high-quality user instructions. We propose TaskEval, a multi-faceted evaluation methodology that assesses LLM performance across these three stages. Our approach combines automated construction with rigorous human verification, ensuring high consistency with human evaluation. Experimental results demonstrate that TaskBench effectively reflects the capabilities of various LLMs in task automation. It provides insights into model performance across different task complexities and domains, pushing the boundaries of what current models can achieve. TaskBench offers a scalable, adaptable, and reliable benchmark for advancing LLM-based autonomous agents.

NeurIPS Conference 2024 Conference Paper

Towards Editing Time Series

• Baoyu Jing
• Shuqi Gu
• Tianyu Chen
• Zhiyu Yang
• Dongsheng Li
• Jingrui He
• Kan Ren

Synthesizing time series data is pivotal in modern society, aiding effective decision making and ensuring privacy preservation in various scenarios. Time series are associated with various attributes, including trends, seasonality, and external information such as location. Recent research has predominantly focused on random unconditional synthesis or conditional synthesis. Nonetheless, these paradigms generate time series from scratch and are incapable of manipulating existing time series samples. This paper introduces a novel task, called Time Series Editing (TSE), to synthesize time series by manipulating existing time series. The objective is to modify the given time series according to the specified attributes while preserving other properties unchanged. This task is not trivial due to the inadequacy of data coverage and the intricate relationships between time series and their attributes. To address these issues, we introduce a novel diffusion model, called TEdit. The proposed TEdit is trained using a novel bootstrap learning algorithm that effectively enhances the coverage of the original data. It is also equipped with an innovative multi-resolution modeling and generation paradigm to capture the complex relationships between time series and their attributes. Experimental results demonstrate the efficacy of TEdit for editing specified attributes upon the existing time series data. The project page is at https: //seqml. github. io/tse.

NeurIPS Conference 2024 Conference Paper

Understanding and Improving Training-free Loss-based Diffusion Guidance

• Yifei Shen
• Xinyang Jiang
• Yifan Yang
• Yezhen Wang
• Dongqi Han
• Dongsheng Li

Adding additional guidance to pretrained diffusion models has become an increasingly popular research area, with extensive applications in computer vision, reinforcement learning, and AI for science. Recently, several studies have proposed training-free loss-based guidance by using off-the-shelf networks pretrained on clean images. This approach enables zero-shot conditional generation for universal control formats, which appears to offer a free lunch in diffusion guidance. In this paper, we aim to develop a deeper understanding of training-free guidance, as well as overcome its limitations. We offer a theoretical analysis that supports training-free guidance from the perspective of optimization, distinguishing it from classifier-based (or classifier-free) guidance. To elucidate their drawbacks, we theoretically demonstrate that training-free guidance is more susceptible to misaligned gradients and exhibits slower convergence rates compared to classifier guidance. We then introduce a collection of techniques designed to overcome the limitations, accompanied by theoretical rationale and empirical evidence. Our experiments in image and motion generation confirm the efficacy of these techniques.

IJCAI Conference 2023 Conference Paper

A Canonicalization-Enhanced Known Fact-Aware Framework For Open Knowledge Graph Link Prediction

• Yilin Wang
• Minghao Hu
• Zhen Huang
• Dongsheng Li
• Wei Luo
• Dong Yang
• Xicheng Lu

Open knowledge graph (OpenKG) link prediction aims to predict missing factual triples in the form of (head noun phrase, relation phrase, tail noun phrase). Since triples are not canonicalized, previous methods either focus on canonicalizing noun phrases (NPs) to reduce graph sparsity, or utilize textual forms to improve type compatibility. However, they neglect to canonicalize relation phrases (RPs) and triples, making OpenKG maintain high sparsity and impeding the performance. To address the above issues, we propose a Canonicalization-Enhanced Known Fact-Aware (CEKFA) framework that boosts link prediction performance through sparsity reduction of RPs and triples. First, we propose a similarity-driven RP canonicalization method to reduce RPs' sparsity by sharing knowledge of semantically similar ones. Second, to reduce the sparsity of triples, a known fact-aware triple canonicalization method is designed to retrieve relevant known facts from training data. Finally, these two types of canonical information are integrated into a general two-stage re-ranking framework that can be applied to most existing knowledge graph embedding methods. Experiment results on two OpenKG datasets, ReVerb20K and ReVerb45K, show that our approach achieves state-of-the-art results. Extensive experimental analyses illustrate the effectiveness and generalization ability of the proposed framework.

AAAI Conference 2023 Conference Paper

Adaptive Policy Learning for Offline-to-Online Reinforcement Learning

• Han Zheng
• Xufang Luo
• Pengfei Wei
• Xuan Song
• Dongsheng Li
• Jing Jiang

Conventional reinforcement learning (RL) needs an environment to collect fresh data, which is impractical when online interactions are costly. Offline RL provides an alternative solution by directly learning from the previously collected dataset. However, it will yield unsatisfactory performance if the quality of the offline datasets is poor. In this paper, we consider an offline-to-online setting where the agent is first learned from the offline dataset and then trained online, and propose a framework called Adaptive Policy Learning for effectively taking advantage of offline and online data. Specifically, we explicitly consider the difference between the online and offline data and apply an adaptive update scheme accordingly, that is, a pessimistic update strategy for the offline dataset and an optimistic/greedy update scheme for the online dataset. Such a simple and effective method provides a way to mix the offline and online RL and achieve the best of both worlds. We further provide two detailed algorithms for implementing the framework through embedding value or policy-based RL algorithms into it. Finally, we conduct extensive experiments on popular continuous control tasks, and results show that our algorithm can learn the expert policy with high sample efficiency even when the quality of offline dataset is poor, e.g., random dataset.

NeurIPS Conference 2023 Conference Paper

ContiFormer: Continuous-Time Transformer for Irregular Time Series Modeling

• Yuqi Chen
• Kan Ren
• Yansen Wang
• Yuchen Fang
• Weiwei Sun
• Dongsheng Li

Modeling continuous-time dynamics on irregular time series is critical to account for data evolution and correlations that occur continuously. Traditional methods including recurrent neural networks or Transformer models leverage inductive bias via powerful neural architectures to capture complex patterns. However, due to their discrete characteristic, they have limitations in generalizing to continuous-time data paradigms. Though neural ordinary differential equations (Neural ODEs) and their variants have shown promising results in dealing with irregular time series, they often fail to capture the intricate correlations within these sequences. It is challenging yet demanding to concurrently model the relationship between input data points and capture the dynamic changes of the continuous-time system. To tackle this problem, we propose ContiFormer that extends the relation modeling of vanilla Transformer to the continuous-time domain, which explicitly incorporates the modeling abilities of continuous dynamics of Neural ODEs with the attention mechanism of Transformers. We mathematically characterize the expressive power of ContiFormer and illustrate that, by curated designs of function hypothesis, many Transformer variants specialized in irregular time series modeling can be covered as a special case of ContiFormer. A wide range of experiments on both synthetic and real-world datasets have illustrated the superior modeling capacities and prediction performance of ContiFormer on irregular time series data. The project link is https: //seqml. github. io/contiformer/.

NeurIPS Conference 2023 Conference Paper

HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face

• Yongliang Shen
• Kaitao Song
• Xu Tan
• Dongsheng Li
• Weiming Lu
• Yueting Zhuang

Solving complicated AI tasks with different domains and modalities is a key step toward artificial general intelligence. While there are numerous AI models available for various domains and modalities, they cannot handle complicated AI tasks autonomously. Considering large language models (LLMs) have exhibited exceptional abilities in language understanding, generation, interaction, and reasoning, we advocate that LLMs could act as a controller to manage existing AI models to solve complicated AI tasks, with language serving as a generic interface to empower this. Based on this philosophy, we present HuggingGPT, an LLM-powered agent that leverages LLMs (e. g. , ChatGPT) to connect various AI models in machine learning communities (e. g. , Hugging Face) to solve AI tasks. Specifically, we use ChatGPT to conduct task planning when receiving a user request, select models according to their function descriptions available in Hugging Face, execute each subtask with the selected AI model, and summarize the response according to the execution results. By leveraging the strong language capability of ChatGPT and abundant AI models in Hugging Face, HuggingGPT can tackle a wide range of sophisticated AI tasks spanning different modalities and domains and achieve impressive results in language, vision, speech, and other challenging tasks, which paves a new way towards the realization of artificial general intelligence.

AAAI Conference 2023 Conference Paper

Learning Decomposed Spatial Relations for Multi-Variate Time-Series Modeling

• Yuchen Fang
• Kan Ren
• Caihua Shan
• Yifei Shen
• You Li
• Weinan Zhang
• Yong Yu
• Dongsheng Li

Modeling multi-variate time-series (MVTS) data is a long-standing research subject and has found wide applications. Recently, there is a surge of interest in modeling spatial relations between variables as graphs, i.e., first learning one static graph for each dataset and then exploiting the graph structure via graph neural networks. However, as spatial relations may differ substantially across samples, building one static graph for all the samples inherently limits flexibility and severely degrades the performance in practice. To address this issue, we propose a framework for fine-grained modeling and utilization of spatial correlation between variables. By analyzing the statistical properties of real-world datasets, a universal decomposition of spatial correlation graphs is first identified. Specifically, the hidden spatial relations can be decomposed into a prior part, which applies across all the samples, and a dynamic part, which varies between samples, and building different graphs is necessary to model these relations. To better coordinate the learning of the two relational graphs, we propose a min-max learning paradigm that not only regulates the common part of different dynamic graphs but also guarantees spatial distinguishability among samples. The experimental results show that our proposed model outperforms the state-of-the-art baseline methods on both time-series forecasting and time-series point prediction tasks.

NeurIPS Conference 2023 Conference Paper

Learning Topology-Agnostic EEG Representations with Geometry-Aware Modeling

• Ke Yi
• Yansen Wang
• Kan Ren
• Dongsheng Li

Large-scale pre-training has shown great potential to enhance models on downstream tasks in vision and language. Developing similar techniques for scalp electroencephalogram (EEG) is suitable since unlabelled data is plentiful. Meanwhile, various sampling channel selections and inherent structural and spatial information bring challenges and avenues to improve existing pre-training strategies further. In order to break boundaries between different EEG resources and facilitate cross-dataset EEG pre-training, we propose to map all kinds of channel selections to a unified topology. We further introduce MMM, a pre-training framework with Multi-dimensional position encoding, Multi-level channel hierarchy, and Multi-stage pre-training strategy built on the unified topology to obtain topology-agnostic representations. Experiments demonstrate that our approach yields impressive improvements over previous state-of-the-art techniques on emotional recognition benchmark datasets.

IJCAI Conference 2023 Conference Paper

Meta-Tsallis-Entropy Minimization: A New Self-Training Approach for Domain Adaptation on Text Classification

• Menglong Lu
• Zhen Huang
• Zhiliang Tian
• Yunxiang Zhao
• Xuanyu Fei
• Dongsheng Li

Text classification is a fundamental task for natural language processing, and adapting text classification models across domains has broad applications. Self-training generates pseudo-examples from the model's predictions and iteratively trains on the pseudo-examples, i. e. , minimizes the loss on the source domain and the Gibbs entropy on the target domain. However, Gibbs entropy is sensitive to prediction errors, and thus, self-training tends to fail when the domain shift is large. In this paper, we propose Meta-Tsallis Entropy minimization (MTEM). MTEM uses an instance adaptive Tsallis entropy to replace the Gibbs entropy and a meta-learning algorithm to optimize the instance adaptive Tsallis entropy on the target domain. To reduce the computation cost of MTEM, we propose an approximation technique to approximate the second-order derivation involved in the meta-learning. To efficiently generate pseudo labels, we propose an annealing sampling mechanism for exploring the model's prediction probability. Theoretically, we prove the convergence of the meta-learning algorithm in MTEM and analyze the effectiveness of MTEM in achieving domain adaptation. Experimentally, MTEM improves the adaptation performance of BERT with an average of 4 percent on the benchmark dataset.

TMLR Journal 2023 Journal Article

Pairwise Learning with Adaptive Online Gradient Descent

• Tao Sun
• Qingsong Wang
• Yunwen Lei
• Dongsheng Li
• Bao Wang

In this paper, we propose an adaptive online gradient descent method with momentum for pairwise learning, in which the stepsize is determined by historical information. Due to the structure of pairwise learning, the sample pairs are dependent on the parameters, causing difficulties in the convergence analysis. To this end, we develop novel techniques for the convergence analysis of the proposed algorithm. We show that the proposed algorithm can output the desired solution in strongly convex, convex, and nonconvex cases. Furthermore, we present theoretical explanations for why our proposed algorithm can accelerate previous workhorses for online pairwise learning. All assumptions used in the theoretical analysis are mild and common, making our results applicable to various pairwise learning problems. To demonstrate the efficiency of our algorithm, we compare the proposed adaptive method with the non-adaptive counterpart on the benchmark online AUC maximization problem.

AAAI Conference 2023 Conference Paper

Similarity Distribution Based Membership Inference Attack on Person Re-identification

• Junyao Gao
• Xinyang Jiang
• Huishuai Zhang
• Yifan Yang
• Shuguang Dou
• Dongsheng Li
• Duoqian Miao
• Cheng Deng

While person Re-identification (Re-ID) has progressed rapidly due to its wide real-world applications, it also causes severe risks of leaking personal information from training data. Thus, this paper focuses on quantifying this risk by membership inference (MI) attack. Most of the existing MI attack algorithms focus on classification models, while Re-ID follows a totally different training and inference paradigm. Re-ID is a fine-grained recognition task with complex feature embedding, and model outputs commonly used by existing MI like logits and losses are not accessible during inference. Since Re-ID focuses on modelling the relative relationship between image pairs instead of individual semantics, we conduct a formal and empirical analysis which validates that the distribution shift of the inter-sample similarity between training and test set is a critical criterion for Re-ID membership inference. As a result, we propose a novel membership inference attack method based on the inter-sample similarity distribution. Specifically, a set of anchor images are sampled to represent the similarity distribution conditioned on a target image, and a neural network with a novel anchor selection module is proposed to predict the membership of the target image. Our experiments validate the effectiveness of the proposed approach on both the Re-ID task and conventional classification task.

AAAI Conference 2023 Conference Paper

Stability-Based Generalization Analysis of the Asynchronous Decentralized SGD

• Xiaoge Deng
• Tao Sun
• Shengwei Li
• Dongsheng Li

The generalization ability often determines the success of machine learning algorithms in practice. Therefore, it is of great theoretical and practical importance to understand and bound the generalization error of machine learning algorithms. In this paper, we provide the first generalization results of the popular stochastic gradient descent (SGD) algorithm in the distributed asynchronous decentralized setting. Our analysis is based on the uniform stability tool, where stable means that the learned model does not change much in small variations of the training set. Under some mild assumptions, we perform a comprehensive generalizability analysis of the asynchronous decentralized SGD, including generalization error and excess generalization error bounds for the strongly convex, convex, and non-convex cases. Our theoretical results reveal the effects of the learning rate, training data size, training iterations, decentralized communication topology, and asynchronous delay on the generalization performance of the asynchronous decentralized SGD. We also study the optimization error regarding the objective function values and investigate how the initial point affects the excess generalization error. Finally, we conduct extensive experiments on MNIST, CIFAR-10, CIFAR-100, and Tiny-ImageNet datasets to validate the theoretical findings.

AAAI Conference 2023 Conference Paper

Towards Inference Efficient Deep Ensemble Learning

• Ziyue Li
• Kan Ren
• Yifan Yang
• Xinyang Jiang
• Yuqing Yang
• Dongsheng Li

Ensemble methods can deliver surprising performance gains but also bring significantly higher computational costs, e.g., can be up to 2048X in large-scale ensemble tasks. However, we found that the majority of computations in ensemble methods are redundant. For instance, over 77% of samples in CIFAR-100 dataset can be correctly classified with only a single ResNet-18 model, which indicates that only around 23% of the samples need an ensemble of extra models. To this end, we propose an inference efficient ensemble learning method, to simultaneously optimize for effectiveness and efficiency in ensemble learning. More specifically, we regard ensemble of models as a sequential inference process and learn the optimal halting event for inference on a specific sample. At each timestep of the inference process, a common selector judges if the current ensemble has reached ensemble effectiveness and halt further inference, otherwise filters this challenging sample for the subsequent models to conduct more powerful ensemble. Both the base models and common selector are jointly optimized to dynamically adjust ensemble inference for different samples with various hardness, through the novel optimization goals including sequential ensemble boosting and computation saving. The experiments with different backbones on real-world datasets illustrate our method can bring up to 56% inference cost reduction while maintaining comparable performance to full ensemble, achieving significantly better ensemble utility than other baselines. Code and supplemental materials are available at https://seqml.github.io/irene.

NeurIPS Conference 2023 Conference Paper

Train Faster, Perform Better: Modular Adaptive Training in Over-Parameterized Models

• Yubin Shi
• Yixuan Chen
• Mingzhi Dong
• Xiaochen Yang
• Dongsheng Li
• Yujiang Wang
• Robert Dick
• Qin Lv

Despite their prevalence in deep-learning communities, over-parameterized models convey high demands of computational costs for proper training. This work studies the fine-grained, modular-level learning dynamics of over-parameterized models to attain a more efficient and fruitful training strategy. Empirical evidence reveals that when scaling down into network modules, such as heads in self-attention models, we can observe varying learning patterns implicitly associated with each module's trainability. To describe such modular-level learning capabilities, we introduce a novel concept dubbed modular neural tangent kernel (mNTK), and we demonstrate that the quality of a module's learning is tightly associated with its mNTK's principal eigenvalue $\lambda_{\max}$. A large $\lambda_{\max}$ indicates that the module learns features with better convergence, while those miniature ones may impact generalization negatively. Inspired by the discovery, we propose a novel training strategy termed Modular Adaptive Training (MAT) to update those modules with their $\lambda_{\max}$ exceeding a dynamic threshold selectively, concentrating the model on learning common features and ignoring those inconsistent ones. Unlike most existing training schemes with a complete BP cycle across all network modules, MAT can significantly save computations by its partially-updating strategy and can further improve performance. Experiments show that MAT nearly halves the computational cost of model training and outperforms the accuracy of baselines.

NeurIPS Conference 2022 Conference Paper

Bootstrapped Transformer for Offline Reinforcement Learning

• Kerong Wang
• Hanye Zhao
• Xufang Luo
• Kan Ren
• Weinan Zhang
• Dongsheng Li

Offline reinforcement learning (RL) aims at learning policies from previously collected static trajectory data without interacting with the real environment. Recent works provide a novel perspective by viewing offline RL as a generic sequence generation problem, adopting sequence models such as Transformer architecture to model distributions over trajectories and repurposing beam search as a planning algorithm. However, the training datasets utilized in general offline RL tasks are quite limited and often suffering from insufficient distribution coverage, which could me harmful to training sequence generation models yet has not drawn enough attention in the previous works. In this paper, we propose a novel algorithm named Bootstrapped Transformer, which incorporates the idea of bootstrapping and leverages the learned model to self-generate more offline data to further boost the training of sequence model. We conduct extensive experiments on two offline RL benchmarks and demonstrate that our model can largely remedy the limitations of the existing offline RL training and beat other strong baseline methods. We also analyze the generated pseudo data and the revealed characteristics may shed some light on offline RL training.

NeurIPS Conference 2022 Conference Paper

Finite-Time Analysis of Adaptive Temporal Difference Learning with Deep Neural Networks

• Tao Sun
• Dongsheng Li
• Bao Wang

Temporal difference (TD) learning with function approximations (linear functions or neural networks) has achieved remarkable empirical success, giving impetus to the development of finite-time analysis. As an accelerated version of TD, the adaptive TD has been proposed and proved to enjoy finite-time convergence under the linear function approximation. Existing numerical results have demonstrated the superiority of adaptive algorithms to vanilla ones. Nevertheless, the performance guarantee of adaptive TD with neural network approximation remains widely unknown. This paper establishes the finite-time analysis for the adaptive TD with multi-layer ReLU network approximation whose samples are generated from a Markov decision process. Our established theory shows that if the width of the deep neural network is large enough, the adaptive TD using neural network approximation can find the (optimal) value function with high probabilities under the same iteration complexity as TD in general cases. Furthermore, we show that the adaptive TD using neural network approximation, with the same width and searching area, can achieve theoretical acceleration when the stochastic semi-gradients decay fast.

AAAI Conference 2022 Conference Paper

Invariant Information Bottleneck for Domain Generalization

• Bo Li
• Yifei Shen
• Yezhen Wang
• Wenzhen Zhu
• Colorado Reed
• Dongsheng Li
• Kurt Keutzer
• Han Zhao

Invariant risk minimization (IRM) has recently emerged as a promising alternative for domain generalization. Nevertheless, the loss function is difficult to optimize for nonlinear classifiers and the original optimization objective could fail when pseudo-invariant features and geometric skews exist. Inspired by IRM, in this paper we propose a novel formulation for domain generalization, dubbed invariant information bottleneck (IIB). IIB aims at minimizing invariant risks for nonlinear classifiers and simultaneously mitigating the impact of pseudo-invariant features and geometric skews. Specifically, we first present a novel formulation for invariant causal prediction via mutual information. Then we adopt the variational formulation of the mutual information to develop a tractable loss function for nonlinear classifiers. To overcome the failure modes of IRM, we propose to minimize the mutual information between the inputs and the corresponding representations. IIB significantly outperforms IRM on synthetic datasets, where the pseudo-invariant features and geometric skews occur, showing the effectiveness of proposed formulation in overcoming failure modes of IRM. Furthermore, experiments on DomainBed show that IIB outperforms 13 baselines by 0. 9% on average across 7 real datasets.

IJCAI Conference 2022 Conference Paper

Modeling Precursors for Temporal Knowledge Graph Reasoning via Auto-encoder Structure

• Yifu Gao
• Linhui Feng
• Zhigang Kan
• Yi Han
• Linbo Qiao
• Dongsheng Li

Temporal knowledge graph (TKG) reasoning that infers missing facts in the future is an essential and challenging task. When predicting a future event, there must be a narrative evolutionary process composed of closely related historical facts to support the event's occurrence, namely fact precursors. However, most existing models employ a sequential reasoning process in an auto-regressive manner, which cannot capture precursor information. This paper proposes a novel auto-encoder architecture that introduces a relation-aware graph attention layer into transformer (rGalT) to accommodate inference over the TKG. Specifically, we first calculate the correlation between historical and predicted facts through multiple attention mechanisms along intra-graph and inter-graph dimensions, then constitute these mutually related facts into diverse fact segments. Next, we borrow the translation generation idea to decode in parallel the precursor information associated with the given query, which enables our model to infer future unknown facts by progressively generating graph structures. Experimental results on four benchmark datasets demonstrate that our model outperforms other state-of-the-art methods, and precursor identification provides supporting evidence for prediction.

AAAI Conference 2022 Conference Paper

Neural Piecewise-Constant Delay Differential Equations

• Qunxi Zhu
• Yifei Shen
• Dongsheng Li
• Wei Lin

Continuous-depth neural networks, such as the Neural Ordinary Differential Equations (ODEs), have aroused a great deal of interest from the communities of machine learning and data science in recent years, which bridge the connection between deep neural networks and dynamical systems. In this article, we introduce a new sort of continuous-depth neural network, called the Neural Piecewise-Constant Delay Differential Equations (PCDDEs). Here, unlike the recently proposed framework of the Neural Delay Differential Equations (DDEs), we transform the single delay into the piecewiseconstant delay(s). The Neural PCDDEs with such a transformation, on one hand, inherit the strength of universal approximating capability in Neural DDEs. On the other hand, the Neural PCDDEs, leveraging the contributions of the information from the multiple previous time steps, further promote the modeling capability without augmenting the network dimension. With such a promotion, we show that the Neural PCDDEs do outperform the several existing continuousdepth neural frameworks on the one-dimensional piecewiseconstant delay population dynamics and real-world datasets, including MNIST, CIFAR10, and SVHN.

NeurIPS Conference 2022 Conference Paper

Parameter-free Dynamic Graph Embedding for Link Prediction

• Jiahao Liu
• Dongsheng Li
• Hansu Gu
• Tun Lu
• Peng Zhang
• Ning Gu

Dynamic interaction graphs have been widely adopted to model the evolution of user-item interactions over time. There are two crucial factors when modelling user preferences for link prediction in dynamic interaction graphs: 1) collaborative relationship among users and 2) user personalized interaction patterns. Existing methods often implicitly consider these two factors together, which may lead to noisy user modelling when the two factors diverge. In addition, they usually require time-consuming parameter learning with back-propagation, which is prohibitive for real-time user preference modelling. To this end, this paper proposes FreeGEM, a parameter-free dynamic graph embedding method for link prediction. Firstly, to take advantage of the collaborative relationships, we propose an incremental graph embedding engine to obtain user/item embeddings, which is an Online-Monitor-Offline architecture consisting of an Online module to approximately embed users/items over time, a Monitor module to estimate the approximation error in real time and an Offline module to calibrate the user/item embeddings when the online approximation errors exceed a threshold. Meanwhile, we integrate attribute information into the model, which enables FreeGEM to better model users belonging to some under represented groups. Secondly, we design a personalized dynamic interaction pattern modeller, which combines dynamic time decay with attention mechanism to model user short-term interests. Experimental results on two link prediction tasks show that FreeGEM can outperform the state-of-the-art methods in accuracy while achieving over 36X improvement in efficiency. All code and datasets can be found in https: //github. com/FudanCISL/FreeGEM.

NeurIPS Conference 2022 Conference Paper

Reinforcement Learning with Automated Auxiliary Loss Search

• Tairan He
• Yuge Zhang
• Kan Ren
• Minghuan Liu
• CHE WANG
• Weinan Zhang
• Yuqing Yang
• Dongsheng Li

A good state representation is crucial to solving complicated reinforcement learning (RL) challenges. Many recent works focus on designing auxiliary losses for learning informative representations. Unfortunately, these handcrafted objectives rely heavily on expert knowledge and may be sub-optimal. In this paper, we propose a principled and universal method for learning better representations with auxiliary loss functions, named Automated Auxiliary Loss Search (A2LS), which automatically searches for top-performing auxiliary loss functions for RL. Specifically, based on the collected trajectory data, we define a general auxiliary loss space of size $7. 5 \times 10^{20}$ and explore the space with an efficient evolutionary search strategy. Empirical results show that the discovered auxiliary loss (namely, A2-winner) significantly improves the performance on both high-dimensional (image) and low-dimensional (vector) unseen tasks with much higher efficiency, showing promising generalization ability to different settings and even different benchmark domains. We conduct a statistical analysis to reveal the relations between patterns of auxiliary losses and RL performance.

IJCAI Conference 2022 Conference Paper

Towards Applicable Reinforcement Learning: Improving the Generalization and Sample Efficiency with Policy Ensemble

• Zhengyu Yang
• Kan Ren
• Xufang Luo
• Minghuan Liu
• Weiqing Liu
• Jiang Bian
• Weinan Zhang
• Dongsheng Li

It is challenging for reinforcement learning (RL) algorithms to succeed in real-world applications. Take financial trading as an example, the market information is noisy yet imperfect and the macroeconomic regulation or other factors may shift between training and evaluation, thus it requires both generalization and high sample efficiency for resolving the task. However, directly applying typical RL algorithms can lead to poor performance in such scenarios. To derive a robust and applicable RL algorithm, in this work, we design a simple but effective method named Ensemble Proximal Policy Optimization (EPPO), which learns ensemble policies in an end-to-end manner. Notably, EPPO combines each policy and the policy ensemble organically and optimizes both simultaneously. In addition, EPPO adopts a diversity enhancement regularization over the policy space which helps to generalize to unseen states and promotes exploration. We theoretically prove that EPPO can increase exploration efficacy, and through comprehensive experimental evaluations on various tasks, we demonstrate that EPPO achieves higher efficiency and is robust for real-world applications compared with vanilla policy optimization algorithms and other ensemble methods. Code and supplemental materials are available at https: //seqml. github. io/eppo.

NeurIPS Conference 2022 Conference Paper

Transcormer: Transformer for Sentence Scoring with Sliding Language Modeling

• Kaitao Song
• Yichong Leng
• Xu Tan
• Yicheng Zou
• Tao Qin
• Dongsheng Li

Sentence scoring aims at measuring the likelihood score of a sentence and is widely used in many natural language processing scenarios, like reranking, which is to select the best sentence from multiple candidates. Previous works on sentence scoring mainly adopted either causal language modeling (CLM) like GPT or masked language modeling (MLM) like BERT, which have some limitations: 1) CLM only utilizes unidirectional information for the probability estimation of a sentence without considering bidirectional context, which affects the scoring quality; 2) MLM can only estimate the probability of partial tokens at a time and thus requires multiple forward passes to estimate the probability of the whole sentence, which incurs large computation and time cost. In this paper, we propose \textit{Transcormer} -- a Transformer model with a novel \textit{sliding language modeling} (SLM) for sentence scoring. Specifically, our SLM adopts a triple-stream self-attention mechanism to estimate the probability of all tokens in a sentence with bidirectional context and only requires a single forward pass. SLM can avoid the limitations of CLM (only unidirectional context) and MLM (multiple forward passes) and inherit their advantages, and thus achieve high effectiveness and efficiency in scoring. Experimental results on multiple tasks demonstrate that our method achieves better performance than other language modelings.

NeurIPS Conference 2022 Conference Paper

VRL3: A Data-Driven Framework for Visual Deep Reinforcement Learning

• CHE WANG
• Xufang Luo
• Keith Ross
• Dongsheng Li

We propose VRL3, a powerful data-driven framework with a simple design for solving challenging visual deep reinforcement learning (DRL) tasks. We analyze a number of major obstacles in taking a data-driven approach, and present a suite of design principles, novel findings, and critical insights about data-driven visual DRL. Our framework has three stages: in stage 1, we leverage non-RL datasets (e. g. ImageNet) to learn task-agnostic visual representations; in stage 2, we use offline RL data (e. g. a limited number of expert demonstrations) to convert the task-agnostic representations into more powerful task-specific representations; in stage 3, we fine-tune the agent with online RL. On a set of challenging hand manipulation tasks with sparse reward and realistic visual inputs, compared to the previous SOTA, VRL3 achieves an average of 780% better sample efficiency. And on the hardest task, VRL3 is 1220% more sample efficient (2440% when using a wider encoder) and solves the task with only 10% of the computation. These significant results clearly demonstrate the great potential of data-driven deep reinforcement learning.

NeurIPS Conference 2021 Conference Paper

Recognizing Vector Graphics without Rasterization

• Xinyang Jiang
• Lu Liu
• Caihua Shan
• Yifei Shen
• Xuanyi Dong
• Dongsheng Li

In this paper, we consider a different data format for images: vector graphics. In contrast to raster graphics which are widely used in image recognition, vector graphics can be scaled up or down into any resolution without aliasing or information loss, due to the analytic representation of the primitives in the document. Furthermore, vector graphics are able to give extra structural information on how low-level elements group together to form high level shapes or structures. These merits of graphic vectors have not been fully leveraged in existing methods. To explore this data format, we target on the fundamental recognition tasks: object localization and classification. We propose an efficient CNN-free pipeline that does not render the graphic into pixels (i. e. rasterization), and takes textual document of the vector graphics as input, called YOLaT (You Only Look at Text). YOLaT builds multi-graphs to model the structural and spatial information in vector graphics, and a dual-stream graph neural network is proposed to detect objects from the graph. Our experiments show that by directly operating on vector graphics, YOLaT outperforms raster-graphic based object detection baselines in terms of both average precision and efficiency. Code is available at https: //github. com/microsoft/YOLaT-VectorGraphicsRecognition.

NeurIPS Conference 2021 Conference Paper

Reinforcement Learning Enhanced Explainer for Graph Neural Networks

• Caihua Shan
• Yifei Shen
• Yao Zhang
• Xiang Li
• Dongsheng Li

Graph neural networks (GNNs) have recently emerged as revolutionary technologies for machine learning tasks on graphs. In GNNs, the graph structure is generally incorporated with node representation via the message passing scheme, making the explanation much more challenging. Given a trained GNN model, a GNN explainer aims to identify a most influential subgraph to interpret the prediction of an instance (e. g. , a node or a graph), which is essentially a combinatorial optimization problem over graph. The existing works solve this problem by continuous relaxation or search-based heuristics. But they suffer from key issues such as violation of message passing and hand-crafted heuristics, leading to inferior interpretability. To address these issues, we propose a RL-enhanced GNN explainer, RG-Explainer, which consists of three main components: starting point selection, iterative graph generation and stopping criteria learning. RG-Explainer could construct a connected explanatory subgraph by sequentially adding nodes from the boundary of the current generated graph, which is consistent with the message passing scheme. Further, we design an effective seed locator to select the starting point, and learn stopping criteria to generate superior explanations. Extensive experiments on both synthetic and real datasets show that RG-Explainer outperforms state-of-the-art GNN explainers. Moreover, RG-Explainer can be applied in the inductive setting, demonstrating its better generalization ability.

AAAI Conference 2021 Conference Paper

Scalable and Explainable 1-Bit Matrix Completion via Graph Signal Learning

• Chao Chen
• Dongsheng Li
• Junchi Yan
• Hanchi Huang
• Xiaokang Yang

One-bit matrix completion is an important class of positiveunlabeled (PU) learning problems where the observations consist of only positive examples, e. g. , in top-N recommender systems. For the first time, we show that 1-bit matrix completion can be formulated as the problem of recovering clean graph signals from noise-corrupted signals in hypergraphs. This makes it possible to enjoy recent advances in graph signal learning. Then, we propose the spectral graph matrix completion (SGMC) method, which can recover the underlying matrix in distributed systems by filtering the noisy data in the graph frequency domain. Meanwhile, it can provide microand macro-level explanations by following vertex-frequency analysis. To tackle the computational and memory issue of performing graph signal operations on large graphs, we construct a scalable Nyström algorithm which can efficiently compute orthonormal eigenvectors. Furthermore, we also develop polynomial and sparse frequency filters to remedy the accuracy loss caused by the approximations. We demonstrate the effectiveness of our algorithms on top-N recommendation tasks, and the results on three large-scale real-world datasets show that SGMC can outperform state-of-the-art top-N recommendation algorithms in accuracy while only requiring a small fraction of training time compared to the baselines.

AAAI Conference 2021 Conference Paper

Stability and Generalization of Decentralized Stochastic Gradient Descent

• Tao Sun
• Dongsheng Li
• Bao Wang

The stability and generalization of stochastic gradient-based methods provide valuable insights into understanding the algorithmic performance of machine learning models. As the main workhorse for deep learning, stochastic gradient descent has received a considerable amount of studies. Nevertheless, the community paid little attention to its decentralized variants. In this paper, we provide a novel formulation of the decentralized stochastic gradient descent. Leveraging this formulation together with (non)convex optimization theory, we establish the first stability and generalization guarantees for the decentralized stochastic gradient descent. Our theoretical results are built on top of a few common and mild assumptions and reveal that the decentralization deteriorates the stability of SGD for the first time. We verify our theoretical findings by using a variety of decentralized settings and benchmark machine learning models.

NeurIPS Conference 2019 Conference Paper

General Proximal Incremental Aggregated Gradient Algorithms: Better and Novel Results under General Scheme

• Tao Sun
• Yuejiao Sun
• Dongsheng Li
• Qing Liao

The incremental aggregated gradient algorithm is popular in network optimization and machine learning research. However, the current convergence results require the objective function to be strongly convex. And the existing convergence rates are also limited to linear convergence. Due to the mathematical techniques, the stepsize in the algorithm is restricted by the strongly convex constant, which may make the stepsize be very small (the strongly convex constant may be small). In this paper, we propose a general proximal incremental aggregated gradient algorithm, which contains various existing algorithms including the basic incremental aggregated gradient method. Better and new convergence results are proved even with the general scheme. The novel results presented in this paper, which have not appeared in previous literature, include: a general scheme, nonconvex analysis, the sublinear convergence rates of the function values, much larger stepsizes that guarantee the convergence, the convergence when noise exists, the line search strategy of the proximal incremental aggregated gradient algorithm and its convergence.

IJCAI Conference 2019 Conference Paper

Heavy-ball Algorithms Always Escape Saddle Points

• Tao Sun
• Dongsheng Li
• Zhe Quan
• Hao Jiang
• Shengguo Li
• Yong Dou

Nonconvex optimization algorithms with random initialization have attracted increasing attention recently. It has been showed that many first-order methods always avoid saddle points with random starting points. In this paper, we answer a question: can the nonconvex heavy-ball algorithms with random initialization avoid saddle points? The answer is yes! Direct using the existing proof technique for the heavy-ball algorithms is hard due to that each iteration of the heavy-ball algorithm consists of current and last points. It is impossible to formulate the algorithms as iteration like xk+1= g(xk) under some mapping g. To this end, we design a new mapping on a new space. With some transfers, the heavy-ball algorithm can be interpreted as iterations after this mapping. Theoretically, we prove that heavy-ball gradient descent enjoys larger stepsize than the gradient descent to escape saddle points to escape the saddle point. And the heavy-ball proximal point algorithm is also considered; we also proved that the algorithm can always escape the saddle point.

AAAI Conference 2019 Conference Paper

Non-Ergodic Convergence Analysis of Heavy-Ball Algorithms

• Tao Sun
• Penghang Yin
• Dongsheng Li
• Chun Huang
• Lei Guan
• Hao Jiang

In this paper, we revisit the convergence of the Heavy-ball method, and present improved convergence complexity results in the convex setting. We provide the first non-ergodic O(1/k) rate result of the Heavy-ball algorithm with constant step size for coercive objective functions. For objective functions satisfying a relaxed strongly convex condition, the linear convergence is established under weaker assumptions on the step size and inertial parameter than made in the existing literature. We extend our results to multi-block version of the algorithm with both the cyclic and stochastic update rules. In addition, our results can also be extended to decentralized optimization, where the ergodic analysis is not applicable.

AAAI Conference 2019 Conference Paper

Read + Verify: Machine Reading Comprehension with Unanswerable Questions

• Minghao Hu
• Furu Wei
• Yuxing Peng
• Zhen Huang
• Nan Yang
• Dongsheng Li

Machine reading comprehension with unanswerable questions aims to abstain from answering when no answer can be inferred. In addition to extract answers, previous works usually predict an additional “no-answer” probability to detect unanswerable cases. However, they fail to validate the answerability of the question by verifying the legitimacy of the predicted answer. To address this problem, we propose a novel read-then-verify system, which not only utilizes a neural reader to extract candidate answers and produce noanswer probabilities, but also leverages an answer verifier to decide whether the predicted answer is entailed by the input snippets. Moreover, we introduce two auxiliary losses to help the reader better handle answer extraction as well as noanswer detection, and investigate three different architectures for the answer verifier. Our experiments on the SQuAD 2. 0 dataset show that our system obtains a score of 74. 2 F1 on test set, achieving state-of-the-art results at the time of submission (Aug. 28th, 2018).

AAAI Conference 2018 Conference Paper

Exploring Temporal Preservation Networks for Precise Temporal Action Localization

• Ke Yang
• Peng Qiao
• Dongsheng Li
• Shaohe Lv
• Yong Dou

Temporal action localization is an important task of computer vision. Though a variety of methods have been proposed, it still remains an open question how to predict the temporal boundaries of action segments precisely. Most works use segment-level classifiers to select video segments pre-determined by action proposal or dense sliding windows. However, in order to achieve more precise action boundaries, a temporal localization system should make dense predictions at a fine granularity. A newly proposed work exploits Convolutional-Deconvolutional-Convolutional (CDC) filters to upsample the predictions of 3D ConvNets, making it possible to perform per-frame action predictions and achieving promising performance in terms of temporal action localization. However, CDC network loses temporal information partially due to the temporal downsampling operation. In this paper, we propose an elegant and powerful Temporal Preservation Convolutional (TPC) Network that equips 3D ConvNets with TPC filters. TPC network can fully preserve temporal resolution and downsample the spatial resolution simultaneously, enabling frame-level granularity action localization with minimal loss of time information. TPC network can be trained in an end-to-end manner. Experiment results on public datasets show that TPC network achieves significant improvement in both per-frame action prediction and segment-level temporal action localization.

AAAI Conference 2017 Conference Paper

ERMMA: Expected Risk Minimization for Matrix Approximation-based Recommender Systems

• Dongsheng Li
• Chao Chen
• Qin Lv
• Li Shang
• Stephen Chu
• Hongyuan Zha

Matrix approximation (MA) is one of the most popular techniques in today’s recommender systems. In most MA-based recommender systems, the problem of risk minimization should be defined, and how to achieve minimum expected risk in model learning is one of the most critical problems to recommendation accuracy. This paper addresses the expected risk minimization problem, in which expected risk can be bounded by the sum of optimization error and generalization error. Based on the uniform stability theory, we propose an expected risk minimized matrix approximation method (ER- MMA), which is designed to achieve better tradeoff between optimization error and generalization error in order to reduce the expected risk of the learned MA models. Theoretical analysis shows that ERMMA can achieve lower expected risk bound than existing MA methods. Experimental results on the MovieLens and Netflix datasets demonstrate that ERMMA outperforms six state-of-the-art MA-based recommendation methods in both rating prediction problem and item ranking problem.

AAAI Conference 2017 Conference Paper

GLOMA: Embedding Global Information in Local Matrix Approximation Models for Collaborative Filtering

• Chao Chen
• Dongsheng Li
• Qin Lv
• Junchi Yan
• Li Shang
• Stephen Chu

Recommender systems have achieved great success in recent years, and matrix approximation (MA) is one of the most popular techniques for collaborative filtering (CF) based recommendation. However, a major issue is that MA methods perform poorly at detecting strong localized associations among closely related users and items. Recently, some MA-based CF methods adopt clustering methods to discover meaningful user-item subgroups and perform ensemble on different clusterings to improve the recommendation accuracy. However, ensemble learning suffers from lower efficiency due to the increased overall computation overhead. In this paper, we propose GLOMA, a new clustering-based matrix approximation method, which can embed global information in local matrix approximation models to improve recommendation accuracy. In GLOMA, a MA model is first trained on the entire data to capture global information. The global MA model is then utilized to guide the training of cluster-based local MA models, such that the local models can detect strong localized associations shared within clusters and at the same time preserve global associations shared among all users/items. Evaluation results using MovieLens and Netflix datasets demonstrate that, by integrating global information in local models, GLOMA can outperform five state-of-the-art MA-based CF methods in recommendation accuracy while achieving descent efficiency.

NeurIPS Conference 2017 Conference Paper

Mixture-Rank Matrix Approximation for Collaborative Filtering

• Dongsheng Li
• Chao Chen
• Wei Liu
• Tun Lu
• Ning Gu
• Stephen Chu

Low-rank matrix approximation (LRMA) methods have achieved excellent accuracy among today's collaborative filtering (CF) methods. In existing LRMA methods, the rank of user/item feature matrices is typically fixed, i. e. , the same rank is adopted to describe all users/items. However, our studies show that submatrices with different ranks could coexist in the same user-item rating matrix, so that approximations with fixed ranks cannot perfectly describe the internal structures of the rating matrix, therefore leading to inferior recommendation accuracy. In this paper, a mixture-rank matrix approximation (MRMA) method is proposed, in which user-item ratings can be characterized by a mixture of LRMA models with different ranks. Meanwhile, a learning algorithm capitalizing on iterated condition modes is proposed to tackle the non-convex optimization problem pertaining to MRMA. Experimental studies on MovieLens and Netflix datasets demonstrate that MRMA can outperform six state-of-the-art LRMA-based CF methods in terms of recommendation accuracy.