Author name cluster

Yixuan Li

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

51 papers

2 author rows

TMLR Journal 2026 Journal Article

How Well Can Preference Optimization Generalize Under Noisy Feedback?

Shawn Im
Yixuan Li

As large language models (LLMs) advance their capabilities, aligning these models with human preferences has become crucial. Preference optimization, which trains models to distinguish between preferred and non-preferred responses based on human feedback, has become a crucial component for aligning LLMs. However, most existing works assume noise-free feedback, which is unrealistic due to the inherent errors and inconsistencies in human judgments. This paper addresses the impact of noisy feedback on preference optimization, providing generalization guarantees under these conditions. In particular, we consider noise models that correspond to common real-world sources of noise, such as mislabeling and uncertainty. Unlike traditional analyses that assume convergence, our work focuses on finite-step preference optimization, offering new insights that are more aligned with practical LLM training. We describe how generalization decays with different types of noise across levels of noise rates based on the preference data distribution and number of samples. Our analysis for noisy preference learning applies to a broad family of preference optimization losses such as DPO, IPO, SLiC, etc. Empirical validation on contemporary LLMs confirms the practical relevance of our findings, offering valuable insights for developing AI systems that align with human preferences.

AAAI Conference 2026 Conference Paper

MicLog: Towards Accurate and Efficient LLM-based Log Parsing via Progressive Meta In-Context Learning

Jianbo Yu
Yixuan Li
Hai Xu
Kang Xu
Junjielong Xu
Zhijing Li
Pinjia He
Wanyuan Wang

Log parsing converts semi-structured logs into structured templates, forming a critical foundation for downstream analysis. Traditional syntax and semantic-based parsers often struggle with semantic variations in evolving logs and data scarcity stemming from their limited domain coverage. Recent large language model (LLM)-based parsers leverage in-context learning (ICL) to extract semantics from examples, demonstrating superior accuracy. However, LLM-based parsers face two main challenges: 1) underutilization of ICL capabilities, particularly in dynamic example selection and cross-domain generalization, leading to inconsistent performance; 2) time-consuming and costly LLM querying. To address these challenges, we present MicLog, the first progressive meta in-context learning (ProgMeta-ICL) log parsing framework that combines meta-learning with ICL on small open-source LLMs (i.e., Qwen-2.5-3B). Specifically, MicLog: i) enhances LLMs' ICL capability through a zero-shot to k-shot ProgMeta-ICL paradigm, employing weighted DBSCAN candidate sampling and enhanced BM25 demonstration selection; ii) accelerates parsing via a multi-level pre-query cache that dynamically matches and refines recently parsed templates. Evaluated on Loghub-2.0, MicLog achieves 10.3% higher parsing accuracy than the state-of-the-art parser while reducing parsing time by 42.4%.

PDF Details DOI

AAMAS Conference 2026 Conference Paper

QDepth-VLA: Quantized Depth Prediction as Auxiliary Supervision for Vision-Language-Action Models

Yixuan Li
Yuhui Chen
Mingcai Zhou
Haoran Li

Spatial perception and reasoning are crucial for Vision–Language– Action(VLA)modelstoaccomplishfine-grainedmanipulationtasks. However, existing approaches often lack the ability to understand and reason over the essential 3D structures necessary for precise control. To address this limitation, we propose QDepth-VLA, a general framework that augments VLA models with an auxiliary depth prediction task. A dedicated depth expert is designed to predict quantized latent tokens of depth maps obtained from a VQ-VAE encoder, enablingthemodeltolearndepth-awarerepresentationsthat capture critical geometric cues. Experimental results on the simulation benchmarks and real-world tasks demonstrate that QDepth- VLA yields strong spatial reasoning and competitive performance on manipulation tasks.

AAAI Conference 2025 Conference Paper

AI-generated Image Quality Assessment in Visual Communication

Yu Tian
Yixuan Li
Baoliang Chen
Hanwei Zhu
Shiqi Wang
Sam Kwong

Assessing the quality of artificial intelligence-generated images (AIGIs) plays a crucial role in their application in real-world scenarios. However, traditional image quality assessment (IQA) algorithms primarily focus on low-level visual perception, while existing IQA works on AIGIs overemphasize the generated content itself, neglecting its effectiveness in real-world applications. To bridge this gap, we propose AIGI-VC, a quality assessment database for AI-Generated Images in Visual Communication, which studies the communicability of AIGIs in the advertising field from the perspectives of information clarity and emotional interaction. The dataset consists of 2,500 images spanning 14 advertisement topics and 8 emotion types. It provides coarse-grained human preference annotations and fine-grained preference descriptions, benchmarking the abilities of IQA methods in preference prediction, interpretation, and reasoning. We conduct an empirical study of existing representative IQA methods and large multi-modal models on the AIGI-VC dataset, uncovering their strengths and weaknesses.

PDF Details DOI

TMLR Journal 2025 Journal Article

CodeLutra: Boosting LLM Code Generation via Preference-Guided Refinement

Leitian Tao
Xiang Chen
Tong Yu
Tung Mai
Ryan A. Rossi
Yixuan Li
Saayan Mitra

Large Language Models (LLMs) have revolutionized code generation but are require significant resources and tend to over-generalize, limiting their task-specific efficiency. Fine-tuning smaller, open-source LLMs is a cost-effective alternative, yet standard supervised approaches rely solely on correct examples, overlooking valuable insights from failures. We introduce CodeLutra, a new framework that leverages both correct and incorrect code attempts. Instead of purely instructing with correct solutions, CodeLutra uses iterative preference-based refinement, comparing successful and failed outputs to better approximate desired results. This process narrows the performance gap with state-of-the-art, larger models, without requiring massive datasets or auxiliary models. For example, on a challenging data science coding task, using only 500 samples improved Llama-3-8B’s accuracy from 28.2% to 48.6%, approaching GPT-4’s level. By capitalizing on both successes and mistakes, \textsc{CodeLutra} offers a scalable, efficient path to high-quality code generation, making smaller open-source models more competitive with leading closed-source alternatives.

NeurIPS Conference 2025 Conference Paper

Direct Numerical Layout Generation for 3D Indoor Scene Synthesis via Spatial Reasoning

Xingjian Ran
Yixuan Li
Linning Xu
Mulin Yu
Bo Dai

Realistic 3D indoor scene synthesis is vital for embodied AI and digital content creation. It can be naturally divided into two subtasks: object generation and layout generation. While recent generative models have significantly advanced object-level quality and controllability, layout generation remains challenging due to limited datasets. Existing methods either overfit to these datasets or rely on predefined constraints to optimize numerical layout that sacrifice flexibility. As a result, they fail to generate scenes that are both open-vocabulary and aligned with fine-grained user instructions. We introduce DirectLayout, a framework that directly generates numerical 3D layouts from text descriptions using generalizable spatial reasoning of large language models (LLMs). DirectLayout decomposes the generation into three stages: producing a Bird's-Eye View (BEV) layout, lifting it into 3D space, and refining object placements. To enable explicit spatial reasoning and help the model grasp basic principles of object placement, we employ Chain-of-Thought (CoT) Activation based on the 3D-Front dataset. Additionally, we design CoT-Grounded Generative Layout Reward to enhance generalization and spatial planning. During inference, DirectLayout addresses asset-layout mismatches via Iterative Asset-Layout Alignment through in-context learning. Extensive experiments demonstrate that DirectLayout achieves impressive semantic consistency, generalization and physical plausibility.

IROS Conference 2025 Conference Paper

Env-Mani: Quadrupedal Robot Loco-Manipulation with Environment-in-the-Loop

Yixuan Li
Zan Wang
Wei Liang

Dogs can climb onto tables using their front legs for support, enabling them to retrieve objects and significantly expand their workspace by leveraging the external environment. However, the ability of quadrupedal robots to perform similar skills remains largely unexplored. In this work, we introduce a unified, learning-based loco-manipulation framework for quadrupedal robots, allowing them to utilize the external environment as support to extend their workspace and enhance their manipulation capabilities. Specifically, our method proposes a unified policy that takes limited onboard sensors and proprioception as input, generating whole-body actions that enable the robot to manipulate objects. To guide the policy learning for environment-in-the-loop manipulation, we design a set of rewards that address challenges such as imprecise perception and center-of-mass shifts. Additionally, we employ curriculum learning to train both teacher and student policies, ensuring effective skill transfer in complex tasks. We train the policy in simulation and conduct extensive experiments, demonstrating that our approach allows robots to manipulate previously inaccessible objects, opening up new possibilities for enhancing quadrupedal robot capabilities without the need for hardware modifications or additional costs. The project page is available at https://sites.google.com/view/env-mani.

AAAI Conference 2025 Conference Paper

Fast and Interpretable Mixed-Integer Linear Program Solving by Learning Model Reduction

Yixuan Li
Can Chen
Jiajun Li
Jiahui Duan
Xiongwei Han
Tao Zhong
Vincent Chau
Weiwei Wu

By exploiting the correlation between the structure and the solution of Mixed-Integer Linear Programming (MILP), Machine Learning (ML) has become a promising method for solving large-scale MILP problems. Existing ML-based MILP solvers mainly focus on end-to-end solution learning, which suffers from the scalability issue due to the high dimensionality of the solution space. Instead of directly learning the optimal solution, this paper aims to learn a reduced and equivalent model of the original MILP as an intermediate step. The reduced model often corresponds to interpretable operations and is much simpler, enabling us to solve large-scale MILP problems much faster than existing commercial solvers. However, current approaches rely only on the optimal reduced model, overlooking the significant preference information of all reduced models. To address this issue, this paper proposes a preference-based model reduction learning method, which considers the relative performance (i.e., objective cost and constraint feasibility) of all reduced models on each MILP instance as preferences. We also introduce an attention mechanism to capture and represent preference information, which helps improve the performance of model reduction learning tasks. Moreover, we propose a SetCover based pruning method to control the number of reduced models (i.e., labels), thereby simplifying the learning process. Evaluation on real-world MILP problems shows that 1) compared to the state-of-the-art model reduction ML methods, our method obtains nearly 20% improvement on solution accuracy, and 2) compared to the commercial solver Gurobi, two to four orders of magnitude speedups are achieved.

PDF Details DOI

TMLR Journal 2025 Journal Article

Generalized Out-of-Distribution Detection and Beyond in Vision Language Model Era: A Survey

Atsuyuki Miyai
Jingkang Yang
Jingyang Zhang
Yifei Ming
Yueqian Lin
Qing Yu
Go Irie
Shafiq Joty

Detecting out-of-distribution (OOD) samples is crucial for ensuring the safety of machine learning systems and has shaped the field of OOD detection. Meanwhile, several other problems are closely related to OOD detection, including anomaly detection (AD), novelty detection (ND), open set recognition (OSR), and outlier detection (OD). To unify these problems, a generalized OOD detection framework was proposed, taxonomically categorizing these five problems. However, Vision Language Models (VLMs) such as CLIP have significantly changed the paradigm and blurred the boundaries between these fields, again confusing researchers. In this survey, we first present a generalized OOD detection v2, encapsulating the evolution of these fields in the VLM era. Our framework reveals that, with some field inactivity and integration, the demanding challenges have become OOD detection and AD. Then, we highlight the significant shift in the definition, problem settings, and benchmarks; we thus feature a comprehensive review of the methodology for OOD detection and related tasks to clarify their relationship to OOD detection. Finally, we explore the advancements in the emerging Large Vision Language Model (LVLM) era, such as GPT-4V. We conclude with open challenges and future directions. The resource is available at https://github.com/AtsuMiyai/Awesome-OOD-VLM.

TMLR Journal 2025 Journal Article

HalluEntity: Benchmarking and Understanding Entity-Level Hallucination Detection

Min-Hsuan Yeh
Max Kamachee
Seongheon Park
Yixuan Li

To mitigate the impact of hallucination nature of LLMs, many studies propose detecting hallucinated generation through uncertainty estimation. However, these approaches predominantly operate at the sentence or paragraph level, failing to pinpoint specific spans or entities responsible for hallucinated content. This lack of granularity is especially problematic for long-form outputs that mix accurate and fabricated information. To address this limitation, we explore entity-level hallucination detection. We propose a new data set, HalluEntity, which annotates hallucination at the entity level. Based on the dataset, we comprehensively evaluate uncertainty-based hallucination detection approaches across 17 modern LLMs. Our experimental results show that uncertainty estimation approaches focusing on individual token probabilities tend to over-predict hallucinations, while context-aware methods show better but still suboptimal performance. Through an in-depth qualitative study, we identify relationships between hallucination tendencies and linguistic properties and highlight important directions for future research. HalluEntity: https://huggingface.co/datasets/samuelyeh/HalluEntity

ICML Conference 2025 Conference Paper

Improving LLM Video Understanding with 16 Frames Per Second

Yixuan Li
Changli Tang
Jimin Zhuang
Yudong Yang
Guangzhi Sun
Wei Li 0119
Zejun Ma 0001
Chao Zhang 0031

Human vision is dynamic and continuous. However, in video understanding with multimodal large language models (LLMs), existing methods primarily rely on static features extracted from images sampled at a fixed low frame rate of frame-per-second (FPS) $\leqslant$2, leading to critical visual information loss. In this paper, we introduce F-16, the first multimodal LLM designed for high-frame-rate video understanding. By increasing the frame rate to 16 FPS and compressing visual tokens within each 1-second clip, F-16 efficiently captures dynamic visual features while preserving key semantic information. Experimental results demonstrate that higher frame rates considerably enhance video understanding across multiple benchmarks, providing a new approach to improving video LLMs beyond scaling model size or training data. F-16 achieves state-of-the-art performance among 7-billion-parameter video LLMs on both general and fine-grained video understanding benchmarks, such as Video-MME and TemporalBench. Furthermore, F-16 excels in complex spatiotemporal tasks, including high-speed sports analysis ( e. g. , basketball, football, gymnastics, and diving), outperforming SOTA proprietary visual models like GPT-4o and Gemini-1. 5-pro. Additionally, we introduce a novel decoding method for F-16 that enables highly efficient low-frame-rate inference without requiring model retraining. We will release the source code, model checkpoints, and data at https: //github. com/bytedance/F-16.

NeurIPS Conference 2025 Conference Paper

Learning Sparse Approximate Inverse Preconditioners for Conjugate Gradient Solvers on GPUs

Zhehao Li
Kangbo Lyu
Yixuan Li
Tao Du
Ligang Liu

The conjugate gradient solver (CG) is a prevalent method for solving symmetric and positive definite linear systems $\mathbf{Ax} = \mathbf{b}$, where effective preconditioners are crucial for fast convergence. Traditional preconditioners rely on prescribed algorithms to offer rigorous theoretical guarantees, while limiting their ability to exploit optimization from data. Existing learning-based methods often utilize Graph Neural Networks (GNNs) to improve the performance and speed up the construction. However, their reliance on incomplete factorization leads to significant challenges: the associated triangular solve hinders GPU parallelization in practice, and introduces long-range dependencies which are difficult for GNNs to model. To address these issues, we propose a learning-based method to generate GPU-friendly preconditioners, particularly using GNNs to construct Sparse Approximate Inverse (SPAI) preconditioners, which avoids triangular solves and requires only two matrix-vector products at each CG step. The locality of matrix-vector product is compatible with the local propagation mechanism of GNNs. The flexibility of GNNs also allows our approach to be applied in a wide range of scenarios. Furthermore, we introduce a statistics-based scale-invariant loss function. Its design matches CG's property that the convergence rate depends on the condition number, rather than the absolute scale of $\mathbf{A}$, leading to improved performance of the learned preconditioner. Evaluations on three PDE-derived datasets and one synthetic dataset demonstrate that our method outperforms standard preconditioners (Diagonal, IC, and traditional SPAI) and previous learning-based preconditioners on GPUs. We reduce solution time on GPUs by 40%-53% (68%-113% faster), along with better condition numbers and superior generalization performance.

AAAI Conference 2025 Conference Paper

Online Prompt Selection for Program Synthesis

Yixuan Li
Lewis Frampton
Federico Mora
Elizabeth Polgreen

Large Language Models (LLMs) demonstrate impressive capabilities in the domain of program synthesis. This level of performance is not, however, universal across all tasks, all LLMs and all prompting styles. There are many areas where one LLM dominates, one prompting style dominates, or where calling a symbolic solver is a better choice than an LLM. A key challenge for the user then, is to identify not only when an LLM is the right choice of solver, and the appropriate LLM to call for a given synthesis task, but also the right way to call it. A non-expert user who makes the wrong choice, incurs a cost both in terms of results (number of tasks solved, and the time it takes to solve them) and financial cost, if using a closed-source language model via a commercial API. We frame this choice as an online learning problem. We use a multi-armed bandit algorithm to select which symbolic solver, or LLM and prompt combination to deploy in order to maximize a given reward function (which may prioritize solving time, number of synthesis tasks solved, or financial cost of solving). We implement an instance of this approach, called \name, and evaluate it on synthesis queries from the literature in ranking function synthesis, from the syntax-guided synthesis competition, and fresh, unseen queries generated from SMT problems. Cyanea solves 37.2 % more queries than the best single solver and achieves results within 4 % of the virtual best solver.

PDF Details DOI

TMLR Journal 2025 Journal Article

Out-of-Distribution Learning with Human Feedback

Haoyue Bai
Xuefeng Du
Katie Rainey
Shibin Parameswaran
Yixuan Li

Out-of-distribution (OOD) learning often relies on strong statistical assumptions or predefined OOD data distributions, limiting its effectiveness in real-world deployment for both OOD generalization and detection, especially when human inspection is minimal. This paper introduces a novel framework for OOD learning that integrates human feedback to enhance model adaptation and reliability. Our approach leverages freely available unlabeled data in the wild, which naturally captures environmental test-time OOD distributions under both covariate and semantic shifts. To effectively utilize such data, we propose selectively acquiring human feedback to label a small subset of informative samples. These labeled samples are then used to train both a multi-class classifier and an OOD detector. By incorporating human feedback, our method significantly improves model robustness and precision in handling OOD scenarios. We provide theoretical insights by establishing generalization error bounds for our algorithm. Extensive experiments demonstrate that our approach outperforms state-of-the-art methods by a significant margin. Code is publicly available at https://github.com/HaoyueBaiZJU/ood-hf.

ICLR Conference 2025 Conference Paper

PostEdit: Posterior Sampling for Efficient Zero-Shot Image Editing

Feng Tian
Yixuan Li
Yichao Yan
Shanyan Guan
Yanhao Ge
Xiaokang Yang 0001

In the field of image editing, three core challenges persist: controllability, background preservation, and efficiency. Inversion-based methods rely on time-consuming optimization to preserve the features of the initial images, which results in low efficiency due to the requirement for extensive network inference. Conversely, inversion-free methods lack theoretical support for background similarity, as they circumvent the issue of maintaining initial features to achieve efficiency. As a consequence, none of these methods can achieve both high efficiency and background consistency. To tackle the challenges and the aforementioned disadvantages, we introduce PostEdit, a method that incorporates a posterior scheme to govern the diffusion sampling process. Specifically, a corresponding measurement term related to both the initial features and Langevin dynamics is introduced to optimize the estimated image generated by the given target prompt. Extensive experimental results indicate that the proposed PostEdit achieves state-of-the-art editing performance while accurately preserving unedited regions. Furthermore, the method is both inversion- and training-free, necessitating approximately 1.5 seconds and 18 GB of GPU memory to generate high-quality results.

EAAI Journal 2025 Journal Article

Predicting the ultimate strength of rectangular concrete-filled steel tube columns under eccentric loading using a knowledge-enhanced machine learning framework

Junbin Lou
Yixuan Li
Shicheng Zheng
Qian Feng
Guannan Wang
Rongqiao Xu
Xudong Qian

Rectangular concrete-filled steel tubes (RCFST) are widely used as structural columns in the field of structural engineering due to their exceptional mechanical properties. In practice, RCFST columns often experience eccentric loads, resulting in combined compression and bending. Although several empirical equations describe the mechanical behavior of CFST columns in terms of mechanical mechanism, their predictive ability is still limited by the necessary simplifying assumptions and application scope. This paper presents the development of a knowledge-enhanced machine learning (KeML) framework, with 15 input parameters, to address the limitations of conventional deterministic approaches and existing machine learning approaches, both suffering from low accuracy and robustness. To facilitate this, a comprehensive database consisting of 447 experimental data is established and divided into training and test sets, enabling the training and evaluation of the proposed KeML model. Comparative analyses against original machine learning models reveal that embedding knowledge can improve predictive accuracy and weak robustness, as well as reduce the required number of training data. Furthermore, a sensitivity analysis demonstrates that the cross-section size predominantly influences the ultimate strength of the RCFST columns subjected to eccentric loading. Finally, based on the sensitivity analysis and correlation analysis, a predictive equation for the ultimate strength of RCFST columns is developed using genetic programming.

EAAI Journal 2025 Journal Article

Prediction of the load–moment interaction curve for rectangular concrete-filled steel tube columns using composite multi-fidelity neural network

Yixuan Li
Junbin Lou
Qian Feng
Guannan Wang
Rongqiao Xu

Rectangular concrete-filled steel tubes (RCFSTs) are widely utilized structural solutions due to their exceptional mechanical behavior under combined compression and bending, characterized by the load–moment (N–M) interaction curve. However, with the advancement of high-performance materials, existing codes are inapplicable, and the scarcity of pure bending experimental data constrains the development of general calculation formulae. In this study, a composite multi-fidelity neural network (CMFNN) is proposed to predict the parabolic N–M curve by incorporating a high-fidelity network and an empirical bending strength method. A total of 1005 axial compression tests and 401 eccentric compression tests collected from a comprehensive literature review are employed to train the proposed model. Compared with specifications and a back propagation neural network, the high-fidelity network, which predicts axial strength through the sequential acquisition of concrete and steel coefficients, demonstrates superior accuracy, with a mean absolute percentage error (MAPE) of 6. 04%. To capture the curve's transition with slenderness, the eccentric compression data is subdivided into three subsets. This approach significantly improves prediction performance, especially in high-slenderness tests. Each subset exhibits a coefficient of determination (R2) above 95% and a MAPE below 8. 72%, outperforming traditional methods. The analysis of the feature importance conducted by the SHapley Additive exPlanations (SHAP) algorithm indicates that the primary factors affecting the curve shape shift from the strength and thickness of steel tube to concrete strength and specimen length as slenderness increases. Additionally, sensitivity analysis demonstrates that height feature predominantly impacts eccentric strength, providing effective guidance for practical engineering applications.

ICML Conference 2025 Conference Paper

video-SALMONN-o1: Reasoning-enhanced Audio-visual Large Language Model

Guangzhi Sun
Yudong Yang
Jimin Zhuang
Changli Tang
Yixuan Li
Wei Li 0119
Zejun Ma 0001
Chao Zhang 0031

While recent advancements in reasoning optimization have significantly enhanced the capabilities of large language models (LLMs), existing efforts to improve reasoning have been limited to solving mathematical problems and focusing on visual graphical inputs, neglecting broader applications in general video understanding. This paper proposes video-SALMONN-o1, the first open-source reasoning-enhanced audio-visual LLM designed for general video understanding tasks. To enhance its reasoning abilities, we develop a reasoning-intensive dataset featuring challenging audio-visual questions with step-by-step solutions. We also propose process direct preference optimization (pDPO), which leverages contrastive step selection to achieve efficient step-level reward modelling tailored for multimodal inputs. Additionally, we introduce RivaBench, the first reasoning-intensive video understanding benchmark, featuring over 4, 000 high-quality, expert-curated question-answer pairs across scenarios such as standup comedy, academic presentations, and synthetic video detection. video-SALMONN-o1 achieves 3-8% accuracy improvements over the LLaVA-OneVision baseline across different video reasoning benchmarks. Besides, pDPO achieves 6-8% improvements compared to the supervised fine-tuning model on RivaBench. Enhanced reasoning enables video-SALMONN-o1 zero-shot synthetic video detection capabilities.

TMLR Journal 2024 Journal Article

A Survey on Out-of-Distribution Detection in NLP

Hao Lang
Yinhe Zheng
Yixuan Li
Jian Sun
Fei Huang
Yongbin Li

Out-of-distribution (OOD) detection is essential for the reliable and safe deployment of machine learning systems in the real world. Great progress has been made over the past years. This paper presents the first review of recent advances in OOD detection with a particular focus on natural language processing approaches. First, we provide a formal definition of OOD detection and discuss several related fields. We then categorize recent algorithms into three classes according to the data they used: (1) OOD data available, (2) OOD data unavailable + in-distribution (ID) label available, and (3) OOD data unavailable + ID label unavailable. Third, we introduce datasets, applications, and metrics. Finally, we summarize existing work and present potential future research topics.

NeurIPS Conference 2024 Conference Paper

Adaptive Image Quality Assessment via Teaching Large Multimodal Model to Compare

Hanwei Zhu
Haoning Wu
Yixuan Li
Zicheng Zhang
Baoliang Chen
Lingyu Zhu
Yuming Fang
Guangtao Zhai

While recent advancements in large multimodal models (LMMs) have significantly improved their abilities in image quality assessment (IQA) relying on absolute quality rating, how to transfer reliable relative quality comparison outputs to continuous perceptual quality scores remains largely unexplored. To address this gap, we introduce an all-around LMM-based NR-IQA model, which is capable of producing qualitatively comparative responses and effectively translating these discrete comparison outcomes into a continuous quality score. Specifically, during training, we present to generate scaled-up comparative instructions by comparing images from the same IQA dataset, allowing for more flexible integration of diverse IQA datasets. Utilizing the established large-scale training corpus, we develop a human-like visual quality comparator. During inference, moving beyond binary choices, we propose a soft comparison method that calculates the likelihood of the test image being preferred over multiple predefined anchor images. The quality score is further optimized by maximum a posteriori estimation with the resulting probability matrix. Extensive experiments on nine IQA datasets validate that the Compare2Score effectively bridges text-defined comparative levels during training with converted single image quality scores for inference, surpassing state-of-the-art IQA models across diverse scenarios. Moreover, we verify that the probability-matrix-based inference conversion not only improves the rating accuracy of Compare2Score but also zero-shot general-purpose LMMs, suggesting its intrinsic effectiveness.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

BackdoorAlign: Mitigating Fine-tuning based Jailbreak Attack with Backdoor Enhanced Safety Alignment

Jiongxiao Wang
Jiazhao Li
Yiquan Li
Xiangyu Qi
Junjie Hu
Yixuan Li
Patrick McDaniel
Muhao Chen

Despite the general capabilities of Large Language Models (LLMs) like GPT-4, these models still request fine-tuning or adaptation with customized data when meeting the specific business demands and intricacies of tailored use cases. However, this process inevitably introduces new safety threats, particularly against the Fine-tuning based Jailbreak Attack (FJAttack) under the setting of Language-Model-as-a-Service (LMaaS), where the model's safety has been significantly compromised by fine-tuning on users' uploaded examples that contain just a few harmful examples. Though potential defenses have been proposed that the service providers of LMaaS can integrate safety examples into the fine-tuning dataset to reduce safety issues, such approaches require incorporating a substantial amount of data, making it inefficient. To effectively defend against the FJAttack with limited safety examples under LMaaS, we propose the Backdoor Enhanced Safety Alignment method inspired by an analogy with the concept of backdoor attacks. In particular, service providers will construct prefixed safety examples with a secret prompt, acting as a "backdoor trigger". By integrating prefixed safety examples into the fine-tuning dataset, the subsequent fine-tuning process effectively acts as the "backdoor attack", establishing a strong correlation between the secret prompt and safety generations. Consequently, safe responses are ensured once service providers prepend this secret prompt ahead of any user input during inference. Our comprehensive experiments demonstrate that through the Backdoor Enhanced Safety Alignment with adding as few as 11 prefixed safety examples, the maliciously fine-tuned LLMs will achieve similar safety performance as the original aligned models without harming the benign performance. Furthermore, we also present the effectiveness of our method in a more practical setting where the fine-tuning data consists of both FJAttack examples and the fine-tuning task data.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

Bridging OOD Detection and Generalization: A Graph-Theoretic View

Han Wang
Yixuan Li

In the context of modern machine learning, models deployed in real-world scenarios often encounter diverse data shifts like covariate and semantic shifts, leading to challenges in both out-of-distribution (OOD) generalization and detection. Despite considerable attention to these issues separately, a unified framework for theoretical understanding and practical usage is lacking. To bridge the gap, we introduce a graph-theoretic framework to jointly tackle both OOD generalization and detection problems. By leveraging the graph formulation, data representations are obtained through the factorization of the graph's adjacency matrix, enabling us to derive provable error quantifying OOD generalization and detection performance. Empirical results showcase competitive performance in comparison to existing methods, thereby validating our theoretical underpinnings.

PDF Details DOI

AAMAS Conference 2024 Conference Paper

Factor Graph Neural Network Meets Max-Sum: A Real-Time Route Planning Algorithm for Massive-Scale Trips

Yixuan Li
Wanyuan Wang
Weiyi XU
Yanchen Deng
Weiwei Wu

Global route planning (GRP) is a typical combinatorial optimization problem that has been solved for a variety of industrial purposes, such as traffic flow management, network routing, and conflict prevention. The goal of the GRP is to find a route for each trip query such that all queries have a minimum global travel time. The GRP problem is NP-hard and computationally challenging, even for medium-sized instances. However, in real-world GRP applications, such as Google Maps-based vehicle route guidance systems, there are always massive-scale trips issued simultaneously, and real-time response is required. Existing mathematical programming-based exact methods and heuristics struggle to balance the extremes of optimality and scalability. Considering that many closed-related GRP instances must be solved repeatedly, this paper explores a deep learning approach to learn real-time and efficient solutions for GRP. This paper first proposes a novel route-query factor graph (RQ-FG) to model the GRP problem, where the message-passing damped Max-sum (DMS) algorithm can be exploited to generate high-quality approximate solutions. A hybrid pruning method is proposed to accelerate solving the DMS. We further devise a route-query factor graph neural network (RQ-FGNN) based on the RQ-FG, which has the ability to return solutions in milliseconds. Experiments demonstrate that our method can generate high-quality solutions in massive-scale GRP instances in real-time.

NeurIPS Conference 2024 Conference Paper

HaloScope: Harnessing Unlabeled LLM Generations for Hallucination Detection

Xuefeng Du
Chaowei Xiao
Yixuan Li

The surge in applications of large language models (LLMs) has prompted concerns about the generation of misleading or fabricated information, known as hallucinations. Therefore, detecting hallucinations has become critical to maintaining trust in LLM-generated content. A primary challenge in learning a truthfulness classifier is the lack of a large amount of labeled truthful and hallucinated data. To address the challenge, we introduce HaloScope, a novel learning framework that leverages the unlabeled LLM generations in the wild for hallucination detection. Such unlabeled data arises freely upon deploying LLMs in the open world, and consists of both truthful and hallucinated information. To harness the unlabeled data, we present an automated scoring function for distinguishing between truthful and untruthful generations within unlabeled mixture data, thereby enabling the training of a binary classifier on top. Importantly, our framework does not require extra data collection and human annotations, offering strong flexibility and practicality for real-world applications. Extensive experiments show that HaloScope can achieve superior hallucination detection performance, outperforming the competitive rivals by a significant margin.

PDF Details DOI

AAAI Conference 2024 Conference Paper

How to Overcome Curse-of-Dimensionality for Out-of-Distribution Detection?

Soumya Suvra Ghosal
Yiyou Sun
Yixuan Li

Machine learning models deployed in the wild can be challenged by out-of-distribution (OOD) data from unknown classes. Recent advances in OOD detection rely on distance measures to distinguish samples that are relatively far away from the in-distribution (ID) data. Despite the promise, distance-based methods can suffer from the curse-of-dimensionality problem, which limits the efficacy in high dimensional feature space. To combat this problem, we propose a novel framework, Subspace Nearest Neighbor (SNN), for OOD detection. In training, our method regularizes the model and its feature representation by leveraging the most relevant subset of dimensions (i.e. subspace). The subspace learning yields highly distinguishable distance measures between ID and OOD data. We provide comprehensive experiments and ablations to validate the efficacy of SNN. Compared to the current best distance-based method, SNN reduces the average FPR95 by 15.96% on the CIFAR-100 benchmark.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

HumanVid: Demystifying Training Data for Camera-controllable Human Image Animation

Zhenzhi Wang
Yixuan Li
Yanhong Zeng
Youqing Fang
Yuwei Guo
Wenran Liu
Jing Tan
Kai Chen

Human image animation involves generating videos from a character photo, allowing user control and unlocking the potential for video and movie production. While recent approaches yield impressive results using high-quality training data, the inaccessibility of these datasets hampers fair and transparent benchmarking. Moreover, these approaches prioritize 2D human motion and overlook the significance of camera motions in videos, leading to limited control and unstable video generation. To demystify the training data, we present HumanVid, the first large-scale high-quality dataset tailored for human image animation, which combines crafted real-world and synthetic data. For the real-world data, we compile a vast collection of real-world videos from the internet. We developed and applied careful filtering rules to ensure video quality, resulting in a curated collection of 20K high-resolution (1080P) human-centric videos. Human and camera motion annotation is accomplished using a 2D pose estimator and a SLAM-based method. To expand our synthetic dataset, we collected 10K 3D avatar assets and leveraged existing assets of body shapes, skin textures and clothings. Notably, we introduce a rule-based camera trajectory generation method, enabling the synthetic pipeline to incorporate diverse and precise camera motion annotation, which can rarely be found in real-world data. To verify the effectiveness of HumanVid, we establish a baseline model named CamAnimate, short for Camera-controllable Human Animation, that considers both human and camera motions as conditions. Through extensive experimentation, we demonstrate that such simple baseline training on our HumanVid achieves state-of-the-art performance in controlling both human pose and camera motions, setting a new benchmark. Demo, data and code could be found in the project website: https: //humanvid. github. io/.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

InterControl: Zero-shot Human Interaction Generation by Controlling Every Joint

Zhenzhi Wang
Jingbo Wang
Yixuan Li
Dahua Lin
Bo Dai

Text-conditioned motion synthesis has made remarkable progress with the emergence of diffusion models. However, the majority of these motion diffusion models are primarily designed for a single character and overlook multi-human interactions. In our approach, we strive to explore this problem by synthesizing human motion with interactions for a group of characters of any size in a zero-shot manner. The key aspect of our approach is the adaptation of human-wise interactions as pairs of human joints that can be either in contact or separated by a desired distance. In contrast to existing methods that necessitate training motion generation models on multi-human motion datasets with a fixed number of characters, our approach inherently possesses the flexibility to model human interactions involving an arbitrary number of individuals, thereby transcending the limitations imposed by the training data. We introduce a novel controllable motion generation method, InterControl, to encourage the synthesized motions maintaining the desired distance between joint pairs. It consists of a motion controller and an inverse kinematics guidance module that realistically and accurately aligns the joints of synthesized characters to the desired location. Furthermore, we demonstrate that the distance between joint pairs for human-wise interactions can be generated using an off-the-shelf Large Language Model (LLM). Experimental results highlight the capability of our framework to generate interactions with multiple human characters and its potential to work with off-the-shelf physics-based character simulators. Code is available at https: //github. com/zhenzhiwang/intercontrol.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

Is A Picture Worth A Thousand Words? Delving Into Spatial Reasoning for Vision Language Models

Jiayu Wang
Yifei Ming
Zhenmei Shi
Vibhav Vineet
Xin Wang
Yixuan Li
Neel Joshi

Large language models (LLMs) and vision-language models (VLMs) have demonstrated remarkable performance across a wide range of tasks and domains. Despite this promise, spatial understanding and reasoning—a fundamental component of human cognition—remains under-explored. We propose SpatialEval, a novel benchmark that covers diverse aspects of spatial reasoning such as relationship understanding, navigation, and counting. We conduct a comprehensive evaluation of competitive language and vision-language models. Our findings reveal several counter-intuitive insights that have been overlooked in the literature: (1) Spatial reasoning poses significant challenges where competitive models can fall behind random guessing; (2) Despite additional visual input, VLMs often under-perform compared to their LLM counterparts; (3) When both textual and visual information is available, multi-modal language models become less reliant on visual information if sufficient textual clues are provided. Additionally, we demonstrate that leveraging redundancy between vision and text can significantly enhance model performance. We hope our study will inform the development of multimodal models to improve spatial intelligence and further close the gap with human intelligence. Our code is available at https: //github. com/jiayuww/SpatialEval.

PDF Details DOI

JMLR Journal 2024 Journal Article

On the Learnability of Out-of-distribution Detection

Zhen Fang
Yixuan Li
Feng Liu
Bo Han
Jie Lu

Supervised learning aims to train a classifier under the assumption that training and test data are from the same distribution. To ease the above assumption, researchers have studied a more realistic setting: out-of-distribution (OOD) detection, where test data may come from classes that are unknown during training (i.e., OOD data). Due to the unavailability and diversity of OOD data, good generalization ability is crucial for effective OOD detection algorithms, and corresponding learning theory is still an open problem. To study the generalization of OOD detection, this paper investigates the probably approximately correct (PAC) learning theory of OOD detection that fits the commonly used evaluation metrics in the literature. First, we find a necessary condition for the learnability of OOD detection. Then, using this condition, we prove several impossibility theorems for the learnability of OOD detection under some scenarios. Although the impossibility theorems are frustrating, we find that some conditions of these impossibility theorems may not hold in some practical scenarios. Based on this observation, we next give several necessary and sufficient conditions to characterize the learnability of OOD detection in some practical scenarios. Lastly, we offer theoretical support for representative OOD detection works based on our OOD theory. [abs] [ pdf ][ bib ] &copy JMLR 2024. ( edit, beta )

ICML Conference 2024 Conference Paper

PICLe: Eliciting Diverse Behaviors from Large Language Models with Persona In-Context Learning

Hyeong Kyu Choi
Yixuan Li

Large Language Models (LLMs) are trained on massive text corpora, which are encoded with diverse personality traits. This triggers an interesting goal of eliciting a desired personality trait from the LLM, and probing its behavioral preferences. Accordingly, we formalize the persona elicitation task, aiming to customize LLM behaviors to align with a target persona. We present Persona In-Context Learning (PICLe), a novel persona elicitation framework grounded in Bayesian inference. At the core, PICLe introduces a new ICL example selection criterion based on likelihood ratio, which is designed to optimally guide the model in eliciting a specific target persona. We demonstrate the effectiveness of PICLe through extensive comparisons against baseline methods across three contemporary LLMs. Code is available at https: //github. com/deeplearning-wisc/picle.

TMLR Journal 2024 Journal Article

Your Classifier Can Be Secretly a Likelihood-Based OOD Detector

Jirayu Burapacheep
Yixuan Li

The ability to detect out-of-distribution (OOD) inputs is critical to guarantee the reliability of classification models deployed in an open environment. A fundamental challenge in OOD detection is that a discriminative classifier is typically trained to estimate the posterior probability $p(y|\mathbf{z})$ for class $y$ given an input $\mathbf{z}$, but lacks the explicit likelihood estimation of $p(\mathbf{z})$ ideally needed for OOD detection. While numerous OOD scoring functions have been proposed for classification models, these estimate scores are often heuristic-driven and cannot be rigorously interpreted as likelihood. To bridge the gap, we propose Intrinsic Likelihood (INK), which offers rigorous likelihood interpretation to modern discriminative-based classifiers. Specifically, our proposed INK score operates on the constrained latent embeddings of a discriminative classifier, which are modeled as a mixture of hyperspherical embeddings with constant norm. We draw a novel connection between the hyperspherical distribution and the intrinsic likelihood, which can be effectively optimized in modern neural networks. Extensive experiments on the OpenOOD benchmark empirically demonstrate that INK establishes a new state-of-the-art in a variety of OOD detection setups, including both far-OOD and near-OOD.

NeurIPS Conference 2023 Conference Paper

A Graph-Theoretic Framework for Understanding Open-World Semi-Supervised Learning

Yiyou Sun
Zhenmei Shi
Yixuan Li

Open-world semi-supervised learning aims at inferring both known and novel classes in unlabeled data, by harnessing prior knowledge from a labeled set with known classes. Despite its importance, there is a lack of theoretical foundations for this problem. This paper bridges the gap by formalizing a graph-theoretic framework tailored for the open-world setting, where the clustering can be theoretically characterized by graph factorization. Our graph-theoretic framework illuminates practical algorithms and provides guarantees. In particular, based on our graph formulation, we apply the algorithm called Spectral Open-world Representation Learning (SORL), and show that minimizing our loss is equivalent to performing spectral decomposition on the graph. Such equivalence allows us to derive a provable error bound on the clustering performance for both known and novel classes, and analyze rigorously when labeled data helps. Empirically, SORL can match or outperform several strong baselines on common benchmark datasets, which is appealing for practical usage while enjoying theoretical guarantees.

AAAI Conference 2023 Conference Paper

Distributionally Robust Optimization with Probabilistic Group

Soumya Suvra Ghosal
Yixuan Li

Modern machine learning models may be susceptible to learning spurious correlations that hold on average but not for the atypical group of samples. To address the problem, previous approaches minimize the empirical worst-group risk. Despite the promise, they often assume that each sample belongs to one and only one group, which does not allow expressing the uncertainty in group labeling. In this paper, we propose a novel framework PG-DRO, which explores the idea of probabilistic group membership for distributionally robust optimization. Key to our framework, we consider soft group membership instead of hard group annotations. The group probabilities can be flexibly generated using either supervised learning or zero-shot approaches. Our framework accommodates samples with group membership ambiguity, offering stronger flexibility and generality than the prior art. We comprehensively evaluate PG-DRO on both image classification and natural language processing benchmarks, establishing superior performance.

PDF Details DOI

NeurIPS Conference 2023 Conference Paper

Dream the Impossible: Outlier Imagination with Diffusion Models

Xuefeng Du
Yiyou Sun
Jerry Zhu
Yixuan Li

Utilizing auxiliary outlier datasets to regularize the machine learning model has demonstrated promise for out-of-distribution (OOD) detection and safe prediction. Due to the labor intensity in data collection and cleaning, automating outlier data generation has been a long-desired alternative. Despite the appeal, generating photo-realistic outliers in the high dimensional pixel space has been an open challenge for the field. To tackle the problem, this paper proposes a new framework Dream-OOD, which enables imagining photo-realistic outliers by way of diffusion models, provided with only the in-distribution (ID) data and classes. Specifically, Dream-OOD learns a text-conditioned latent space based on ID data, and then samples outliers in the low-likelihood region via the latent, which can be decoded into images by the diffusion model. Different from prior works [16, 95], Dream-OOD enables visualizing and understanding the imagined outliers, directly in the pixel space. We conduct comprehensive quantitative and qualitative studies to understand the efficacy of Dream-OOD, and show that training with the samples generated by Dream-OOD can significantly benefit OOD detection performance.

IROS Conference 2023 Conference Paper

I2P-Rec: Recognizing Images on Large-Scale Point Cloud Maps Through Bird's Eye View Projections

Shuhang Zheng
Yixuan Li
Zhu Yu 0001
Beinan Yu
Si-Yuan Cao
Minhang Wang
Jintao Xu 0001
Rui Ai 0001

Place recognition is an important technique for autonomous cars to achieve full autonomy since it can provide an initial guess to online localization algorithms. Although current methods based on images or point clouds have achieved satisfactory performance, localizing the images on a large-scale point cloud map remains a fairly unexplored problem. This cross-modal matching task is challenging due to the difficulty in extracting consistent descriptors from images and point clouds. In this paper, we propose the I2P-Rec method to solve the problem by transforming the cross-modal data into the same modality. Specifically, we leverage on the recent success of depth estimation networks to recover point clouds from images. We then project the point clouds into Bird's Eye View (BEV) images. Using the BEV image as an intermediate representation, we extract global features with a Convolutional Neural Network followed by a NetVLAD layer to perform matching. The experimental results evaluated on the KITTI dataset show that, with only a small set of training data, I2P-Rec achieves recall rates at Top-l % over 80% and 90%, when localizing monocular and stereo images on point cloud maps, respectively. We further evaluate I2P-Rec on a 1 km trajectory dataset collected by an autonomous logistics car and show that I2P- Rec can generalize well to previously unseen environments.

NeurIPS Conference 2023 Conference Paper

Learning to Augment Distributions for Out-of-distribution Detection

Qizhou Wang
Zhen Fang
Yonggang Zhang
Feng Liu
Yixuan Li
Bo Han

Open-world classification systems should discern out-of-distribution (OOD) data whose labels deviate from those of in-distribution (ID) cases, motivating recent studies in OOD detection. Advanced works, despite their promising progress, may still fail in the open world, owing to the lacking knowledge about unseen OOD data in advance. Although one can access auxiliary OOD data (distinct from unseen ones) for model training, it remains to analyze how such auxiliary data will work in the open world. To this end, we delve into such a problem from a learning theory perspective, finding that the distribution discrepancy between the auxiliary and the unseen real OOD data is the key to affect the open-world detection performance. Accordingly, we propose Distributional-Augmented OOD Learning (DAOL), alleviating the OOD distribution discrepancy by crafting an OOD distribution set that contains all distributions in a Wasserstein ball centered on the auxiliary OOD distribution. We justify that the predictor trained over the worst OOD data in the ball can shrink the OOD distribution discrepancy, thus improving the open-world detection performance given only the auxiliary OOD data. We conduct extensive evaluations across representative OOD detection setups, demonstrating the superiority of our DAOL over its advanced counterparts.

TMLR Journal 2023 Journal Article

Noise-robust Graph Learning by Estimating and Leveraging Pairwise Interactions

Xuefeng Du
Tian Bian
Yu Rong
Bo Han
Tongliang Liu
Tingyang Xu
Wenbing Huang
Yixuan Li

Teaching Graph Neural Networks (GNNs) to accurately classify nodes under severely noisy labels is an important problem in real-world graph learning applications, but is currently underexplored. Although pairwise training methods have demonstrated promise in supervised metric learning and unsupervised contrastive learning, they remain less studied on noisy graphs, where the structural pairwise interactions (PI) between nodes are abundant and thus might benefit label noise learning rather than the pointwise methods. This paper bridges the gap by proposing a pairwise framework for noisy node classification on graphs, which relies on the PI as a primary learning proxy in addition to the pointwise learning from the noisy node class labels. Our proposed framework PI-GNN contributes two novel components: (1) a confidence-aware PI estimation model that adaptively estimates the PI labels, which are defined as whether the two nodes share the same node labels, and (2) a decoupled training approach that leverages the estimated PI labels to regularize a node classification model for robust node classification. Extensive experiments on different datasets and GNN architectures demonstrate the effectiveness of PI-GNN, yielding a promising improvement over the state-of-the-art methods. Code is publicly available at https://github.com/TianBian95/pi-gnn.

TMLR Journal 2023 Journal Article

OpenCon: Open-world Contrastive Learning

Yiyou Sun
Yixuan Li

Machine learning models deployed in the wild naturally encounter unlabeled samples from both known and novel classes. Challenges arise in learning from both the labeled and unlabeled data, in an open-world semi-supervised manner. In this paper, we introduce a new learning framework, open-world contrastive learning (OpenCon). OpenCon tackles the challenges of learning compact representations for both known and novel classes and facilitates novelty discovery along the way. We demonstrate the effectiveness of OpenCon on challenging benchmark datasets and establish competitive performance. On the ImageNet dataset, OpenCon significantly outperforms the current best method by 11.9% and 7.4% on novel and overall classification accuracy, respectively. Theoretically, OpenCon can be rigorously interpreted from an EM algorithm perspective—minimizing our contrastive loss partially maximizes the likelihood by clustering similar samples in the embedding space. The code is available at https://github.com/deeplearning-wisc/opencon.

TMLR Journal 2022 Journal Article

A Unified Survey on Anomaly, Novelty, Open-Set, and Out of-Distribution Detection: Solutions and Future Challenges

Mohammadreza Salehi
Hossein Mirzaei
Dan Hendrycks
Yixuan Li
Mohammad Hossein Rohban
Mohammad Sabokrou

Machine learning models often encounter samples that are diverged from the training distribution. Failure to recognize an out-of-distribution (OOD) sample, and consequently assign that sample to an in-class label, significantly compromises the reliability of a model. The problem has gained significant attention due to its importance for safety deploying models in open-world settings. Detecting OOD samples is challenging due to the intractability of modeling all possible unknown distributions. To date, several research domains tackle the problem of detecting unfamiliar samples, including anomaly detection, novelty detection, one-class learning, open set recognition, and out-of-distribution detection. Despite having similar and shared concepts, out-of-distribution, open-set, and anomaly detection have been investigated independently. Accordingly, these research avenues have not crosspollinated, creating research barriers. While some surveys intend to provide an overview of these approaches, they seem to only focus on a specific domain without examining the relationship between different domains. This survey aims to provide a cross-domain and comprehensive review of numerous eminent works in respective areas while identifying their commonalities. Researchers can benefit from the overview of research advances in different fields and develop future methodology synergistically. Furthermore, to the best of our knowledge, while there are surveys in anomaly detection or one-class learning, there is no comprehensive or up-to-date survey on out-of-distribution detection, which this survey covers extensively. Finally, having a unified cross-domain perspective, this study discusses and sheds light on future lines of research, intending to bring these fields closer together.

NeurIPS Conference 2022 Conference Paper

Delving into Out-of-Distribution Detection with Vision-Language Representations

Yifei Ming
Ziyang Cai
Jiuxiang Gu
Yiyou Sun
Wei Li
Yixuan Li

Recognizing out-of-distribution (OOD) samples is critical for machine learning systems deployed in the open world. The vast majority of OOD detection methods are driven by a single modality (e. g. , either vision or language), leaving the rich information in multi-modal representations untapped. Inspired by the recent success of vision-language pre-training, this paper enriches the landscape of OOD detection from a single-modal to a multi-modal regime. Particularly, we propose Maximum Concept Matching (MCM), a simple yet effective zero-shot OOD detection method based on aligning visual features with textual concepts. We contribute in-depth analysis and theoretical insights to understand the effectiveness of MCM. Extensive experiments demonstrate that MCM achieves superior performance on a wide variety of real-world tasks. MCM with vision-language features outperforms a common baseline with pure visual features on a hard OOD task with semantically similar classes by 13. 1% (AUROC) Code is available at https: //github. com/deeplearning-wisc/MCM.

NeurIPS Conference 2022 Conference Paper

Is Out-of-Distribution Detection Learnable?

Zhen Fang
Yixuan Li
Jie Lu
Jiahua Dong
Bo Han
Feng Liu

Supervised learning aims to train a classifier under the assumption that training and test data are from the same distribution. To ease the above assumption, researchers have studied a more realistic setting: out-of-distribution (OOD) detection, where test data may come from classes that are unknown during training (i. e. , OOD data). Due to the unavailability and diversity of OOD data, good generalization ability is crucial for effective OOD detection algorithms. To study the generalization of OOD detection, in this paper, we investigate the probably approximately correct (PAC) learning theory of OOD detection, which is proposed by researchers as an open problem. First, we find a necessary condition for the learnability of OOD detection. Then, using this condition, we prove several impossibility theorems for the learnability of OOD detection under some scenarios. Although the impossibility theorems are frustrating, we find that some conditions of these impossibility theorems may not hold in some practical scenarios. Based on this observation, we next give several necessary and sufficient conditions to characterize the learnability of OOD detection in some practical scenarios. Lastly, we also offer theoretical supports for several representative OOD detection works based on our OOD theory.

AAAI Conference 2022 Conference Paper

On the Impact of Spurious Correlation for Out-of-Distribution Detection

Yifei Ming
Hang Yin
Yixuan Li

Modern neural networks can assign high confidence to inputs drawn from outside the training distribution, posing threats to models in real-world deployments. While much research attention has been placed on designing new out-of-distribution (OOD) detection methods, the precise definition of OOD is often left in vagueness and falls short of the desired notion of OOD in reality. In this paper, we present a new formalization and model the data shifts by taking into account both the invariant and environmental (spurious) features. Under such formalization, we systematically investigate how spurious correlation in the training set impacts OOD detection. Our results suggest that the detection performance is severely worsened when the correlation between spurious features and labels is increased in the training set. We further show insights on detection methods that are more effective in reducing the impact of spurious correlation, and provide theoretical analysis on why reliance on environmental features leads to high OOD detection error. Our work aims to facilitate better understandings of OOD samples and their formalization, as well as the exploration of methods that enhance OOD detection. Code is available at https: //github. com/deeplearning-wisc/Spurious_OOD.

NeurIPS Conference 2022 Conference Paper

OpenOOD: Benchmarking Generalized Out-of-Distribution Detection

Jingkang Yang
Pengyun Wang
Dejian Zou
Zitang Zhou
Kunyuan Ding
Wenxuan Peng
Haoqi Wang
Guangyao Chen

Out-of-distribution (OOD) detection is vital to safety-critical machine learning applications and has thus been extensively studied, with a plethora of methods developed in the literature. However, the field currently lacks a unified, strictly formulated, and comprehensive benchmark, which often results in unfair comparisons and inconclusive results. From the problem setting perspective, OOD detection is closely related to neighboring fields including anomaly detection (AD), open set recognition (OSR), and model uncertainty, since methods developed for one domain are often applicable to each other. To help the community to improve the evaluation and advance, we build a unified, well-structured codebase called OpenOOD, which implements over 30 methods developed in relevant fields and provides a comprehensive benchmark under the recently proposed generalized OOD detection framework. With a comprehensive comparison of these methods, we are gratified that the field has progressed significantly over the past few years, where both preprocessing methods and the orthogonal post-hoc methods show strong potential.

AAAI Conference 2022 Conference Paper

Provable Guarantees for Understanding Out-of-Distribution Detection

Peyman Morteza
Yixuan Li

Out-of-distribution (OOD) detection is important for deploying machine learning models in the real world, where test data from shifted distributions can naturally arise. While a plethora of algorithmic approaches have recently emerged for OOD detection, a critical gap remains in theoretical understanding. In this work, we develop an analytical framework that characterizes and unifies the theoretical understanding for OOD detection. Our analytical framework motivates a novel OOD detection method for neural networks, GEM, which demonstrates both theoretical and empirical superiority. In particular, on CIFAR-100 as in-distribution data, our method outperforms a competitive baseline by 16. 57% (FPR95). Lastly, we formally provide provable guarantees and comprehensive analysis of our method, underpinning how various properties of data distribution affect the performance of OOD detection.

NeurIPS Conference 2022 Conference Paper

SIREN: Shaping Representations for Detecting Out-of-Distribution Objects

Xuefeng Du
Gabriel Gozum
Yifei Ming
Yixuan Li

Detecting out-of-distribution (OOD) objects is indispensable for safely deploying object detectors in the wild. Although distance-based OOD detection methods have demonstrated promise in image classification, they remain largely unexplored in object-level OOD detection. This paper bridges the gap by proposing a distance-based framework for detecting OOD objects, which relies on the model-agnostic representation space and provides strong generality across different neural architectures. Our proposed framework SIREN contributes two novel components: (1) a representation learning component that uses a trainable loss function to shape the representations into a mixture of von Mises-Fisher (vMF) distributions on the unit hypersphere, and (2) a test-time OOD detection score leveraging the learned vMF distributions in a parametric or non-parametric way. SIREN achieves competitive performance on both the recent detection transformers and CNN-based models, improving the AUROC by a large margin compared to the previous best method. Code is publicly available at https: //github. com/deeplearning-wisc/siren.

NeurIPS Conference 2022 Conference Paper

SoLar: Sinkhorn Label Refinery for Imbalanced Partial-Label Learning

Haobo Wang
Mingxuan Xia
Yixuan Li
Yuren Mao
Lei Feng
Gang Chen
Junbo Zhao

Partial-label learning (PLL) is a peculiar weakly-supervised learning task where the training samples are generally associated with a set of candidate labels instead of single ground truth. While a variety of label disambiguation methods have been proposed in this domain, they normally assume a class-balanced scenario that may not hold in many real-world applications. Empirically, we observe degenerated performance of the prior methods when facing the combinatorial challenge from the long-tailed distribution and partial-labeling. In this work, we first identify the major reasons that the prior work failed. We subsequently propose SoLar, a novel Optimal Transport-based framework that allows to refine the disambiguated labels towards matching the marginal class prior distribution. SoLar additionally incorporates a new and systematic mechanism for estimating the long-tailed class prior distribution under the PLL setup. Through extensive experiments, SoLar exhibits substantially superior results on standardized benchmarks compared to the previous state-of-the-art PLL methods. Code and data are available at: https: //github. com/hbzju/SoLar.

NeurIPS Conference 2021 Conference Paper

Can multi-label classification networks know what they don’t know?

Haoran Wang
Weitang Liu
Alex Bocchieri
Yixuan Li

Estimating out-of-distribution (OOD) uncertainty is a major challenge for safely deploying machine learning models in the open-world environment. Improved methods for OOD detection in multi-class classification have emerged, while OOD detection methods for multi-label classification remain underexplored and use rudimentary techniques. We propose JointEnergy, a simple and effective method, which estimates the OOD indicator scores by aggregating label-wise energy scores from multiple labels. We show that JointEnergy can be mathematically interpreted from a joint likelihood perspective. Our results show consistent improvement over previous methods that are based on the maximum-valued scores, which fail to capture joint information from multiple labels. We demonstrate the effectiveness of our method on three common multi-label classification benchmarks, including MS-COCO, PASCAL-VOC, and NUS-WIDE. We show that JointEnergy can reduce the FPR95 by up to 10. 05% compared to the previous best baseline, establishing state-of-the-art performance.

NeurIPS Conference 2021 Conference Paper

On the Importance of Gradients for Detecting Distributional Shifts in the Wild

Rui Huang
Andrew Geng
Yixuan Li

Detecting out-of-distribution (OOD) data has become a critical component in ensuring the safe deployment of machine learning models in the real world. Existing OOD detection approaches primarily rely on the output or feature space for deriving OOD scores, while largely overlooking information from the gradient space. In this paper, we present GradNorm, a simple and effective approach for detecting OOD inputs by utilizing information extracted from the gradient space. GradNorm directly employs the vector norm of gradients, backpropagated from the KL divergence between the softmax output and a uniform probability distribution. Our key idea is that the magnitude of gradients is higher for in-distribution (ID) data than that for OOD data, making it informative for OOD detection. GradNorm demonstrates superior performance, reducing the average FPR95 by up to 16. 33% compared to the previous best method.

NeurIPS Conference 2021 Conference Paper

ReAct: Out-of-distribution Detection With Rectified Activations

Yiyou Sun
Chuan Guo
Yixuan Li

Out-of-distribution (OOD) detection has received much attention lately due to its practical importance in enhancing the safe deployment of neural networks. One of the primary challenges is that models often produce highly confident predictions on OOD data, which undermines the driving principle in OOD detection that the model should only be confident about in-distribution samples. In this work, we propose ReAct—a simple and effective technique for reducing model overconfidence on OOD data. Our method is motivated by novel analysis on internal activations of neural networks, which displays highly distinctive signature patterns for OOD distributions. Our method can generalize effectively to different network architectures and different OOD detection scores. We empirically demonstrate that ReAct achieves competitive detection performance on a comprehensive suite of benchmark datasets, and give theoretical explication for our method’s efficacy. On the ImageNet benchmark, ReAct reduces the false positive rate (FPR95) by 25. 05% compared to the previous best method.

NeurIPS Conference 2020 Conference Paper

Energy-based Out-of-distribution Detection

Weitang Liu
Xiaoyun Wang
John Owens
Yixuan Li

Determining whether inputs are out-of-distribution (OOD) is an essential building block for safely deploying machine learning models in the open world. However, previous methods relying on the softmax confidence score suffer from overconfident posterior distributions for OOD data. We propose a unified framework for OOD detection that uses an energy score. We show that energy scores better distinguish in- and out-of-distribution samples than the traditional approach using the softmax scores. Unlike softmax confidence scores, energy scores are theoretically aligned with the probability density of the inputs and are less susceptible to the overconfidence issue. Within this framework, energy can be flexibly used as a scoring function for any pre-trained neural classifier as well as a trainable cost function to shape the energy surface explicitly for OOD detection. On a CIFAR-10 pre-trained WideResNet, using the energy score reduces the average FPR (at TPR 95%) by 18. 03% compared to the softmax confidence score. With energy-based training, our method outperforms the state-of-the-art on common benchmarks.

NeurIPS Conference 2020 Conference Paper

Learning Implicit Credit Assignment for Cooperative Multi-Agent Reinforcement Learning

Meng Zhou
Ziyu Liu
Pengwei Sui
Yixuan Li
Yuk Ying Chung

We present a multi-agent actor-critic method that aims to implicitly address the credit assignment problem under fully cooperative settings. Our key motivation is that credit assignment among agents may not require an explicit formulation as long as (1) the policy gradients derived from a centralized critic carry sufficient information for the decentralized agents to maximize their joint action value through optimal cooperation and (2) a sustained level of exploration is enforced throughout training. Under the centralized training with decentralized execution (CTDE) paradigm, we achieve the former by formulating the centralized critic as a hypernetwork such that a latent state representation is integrated into the policy gradients through its multiplicative association with the stochastic policies; to achieve the latter, we derive a simple technique called adaptive entropy regularization where magnitudes of the entropy gradients are dynamically rescaled based on the current policy stochasticity to encourage consistent levels of exploration. Our algorithm, referred to as LICA, is evaluated on several benchmarks including the multi-agent particle environments and a set of challenging StarCraft II micromanagement tasks, and we show that LICA significantly outperforms previous methods.