Jindong Wang

TIST Journal 2026 Journal Article

Evaluating LLM-based Agents for Multi-turn Conversations: A Survey

• Shengyue Guan
• Jindong Wang
• Jiang Bian
• Bin Zhu
• Jian-Guang Lou
• Haoyi Xiong

This survey examines evaluation methods for large language model (LLM)-based agents in multi-turn conversational settings. Using a PRISMA-inspired framework, we systematically reviewed nearly 250 scholarly sources, capturing the state-of-the-art from various venues of publication, and establishing a solid foundation for our analysis. Our study offers a structured approach by developing two interrelated taxonomy systems: one that defines what to evaluate and another that explains how to evaluate. The first taxonomy identifies key components of LLM-based agents for multi-turn conversations and their evaluation dimensions, including task completion, response quality, user experience, memory and context retention, as well as planning and tool integration. These components ensure that the performance of conversational agents is assessed in a holistic and meaningful manner. The second taxonomy system focuses on the evaluation methodologies. It categorizes approaches into annotation-based evaluations, automated metrics, hybrid strategies that combine human assessments with quantitative measures, and self-judging methods utilizing LLMs. This framework not only captures traditional metrics derived from language understanding, such as BLEU and ROUGE scores, but also incorporates advanced techniques that reflect the dynamic, interactive nature of multi-turn dialogues. Together, these frameworks summarize the current status quo, expose limitations in traditional practices, and provide a structured blueprint for improvement. Based on the summarization of existing studies, we identify several challenges and propose future directions, including the development of scalable, real-time evaluation pipelines, enhanced privacy-preserving mechanisms, and robust metrics that capture dynamic multi-turn interactions. Our contributions bridge historical insights with modern practices, paving the way for next-generation, reliably evaluated conversational AI systems and offering a comprehensive guide for researchers and practitioners.

TIST Journal 2026 Journal Article

Introduction to the Special Issue on Evaluations of Large Language Models Part 2

• Jindong Wang
• Linyi Yang
• Sunayana Sitaram
• Qiang Yang
• Bhiksha Raj

AAAI Conference 2026 Conference Paper

Is Your (Reasoning) Multimodal Language Model Vulnerable Toward Distractions?

• Ming Liu
• Hao Chen
• Jindong Wang
• Liwen Wang
• Jingchen Sun
• Wensheng Zhang

Vision-Language Models (VLMs) have achieved success in tasks such as visual question answering, yet their resilience to distractions remains underexplored. Understanding how distractions affect VLMs' performance is crucial for real-world applications, as input data often contains noisy or irrelevant content. This paper assesses the robustness of VLMs—including general-purpose models and those specialized for reasoning—against distractions in the context of science question answering. We introduce I-ScienceQA, a new benchmark based on the ScienceQA dataset, which systematically injects distractions into both visual and textual contexts. We evaluate how distractions perturb the underlying reasoning processes of these models by analyzing changes in textual explanations leading to answers. Our findings show that most VLMs are vulnerable to distractions, with a noticeable degradation in reasoning when extraneous content is present. In particular, some models (including GPT-o4 mini) exhibit a higher degree of robustness. We also observe that textual distractions generally cause greater performance declines than visual distractions. Finally, we explore mitigation strategies such as prompt engineering. Although these strategies improve resilience modestly, our analysis highlights considerable room for further improvement in the robustness of VLMs.

NeurIPS Conference 2025 Conference Paper

From Pretraining to Pathology: How Noise Leads to Catastrophic Inheritance in Medical Models

• Hao Sun
• Zhongyi Han
• Hao Chen
• Jindong Wang
• Xin Gao
• Yilong Yin

Foundation models pretrained on web-scale data drive contemporary transfer learning in vision, language, and multimodal tasks. Recent work shows that mild label noise in these corpora may lift in-distribution accuracy yet sharply reduce out-of-distribution generalization, an effect known as catastrophic inheritance. Medical data is especially sensitive because annotations are scarce, domain shifts are large, and pretraining sources are noisy. We present the first systematic analysis of catastrophic inheritance in medical models. Controlled label-corruption experiments expose a clear structural collapse: as noise rises, the skewness and kurtosis of feature and logit distributions decline, signaling a flattened representation space and diminished discriminative detail. These higher-order statistics form a compact, interpretable marker of degradation in fine-grained tasks such as histopathology. Guided by this finding, we introduce a fine-tuning objective that restores skewness and kurtosis through two scalar regularizers added to the task loss. The method leaves the backbone unchanged and incurs negligible overhead. Tests on PLIP models trained with Twitter pathology images, as well as other large-scale vision and language backbones, show consistent gains in robustness and cross-domain accuracy under varied noise levels.

TIST Journal 2025 Journal Article

Introduction to the Special Issue on Evaluations of Large Language Models: Part 1

• Jindong Wang
• Linyi Yang
• Sunayana Sitaram
• Qiang Yang
• Bhiksha Raj

NeurIPS Conference 2025 Conference Paper

On Fairness of Unified Multimodal Large Language Model for Image Generation

• Ming Liu
• Hao Chen
• Jindong Wang
• Liwen Wang
• Bhiksha Raj
• Wensheng Zhang

Unified multimodal large language models (U-MLLMs) have demonstrated impressive performance in end-to-end visual understanding and generation tasks. However, compared to generation-only systems (e. g. , Stable Diffusion), the unified architecture of U-MLLMs introduces new risks of propagating demographic stereotypes. In this paper, we benchmark several state-of-the-art U-MLLMs and show that they exhibit significant gender and race biases in the generated outputs. To diagnose the source of these biases, we propose a locate-then-fix framework: we first audit the vision and language components — using techniques such as linear probing and controlled generation — and find that the language model appears to be a primary origin of the observed generative bias. Moreover, we observe a ``partial alignment'' phenomenon, where the U-MLLMs exhibit less bias in understanding tasks yet produce substantially biased images. To address this, we introduce a novel \emph{balanced preference loss} that enforces uniform generation probabilities across demographics by leveraging a synthetically balanced dataset. Extensive experiments show that our approach significantly reduces demographic bias while preserving semantic fidelity and image quality. Our findings underscore the need for targeted debiasing strategies in unified multimodal systems and introduce a practical approach to mitigate biases.

NeurIPS Conference 2025 Conference Paper

Personalized Safety in LLMs: A Benchmark and A Planning-Based Agent Approach

• Yuchen Wu
• Edward Sun
• Kaijie Zhu
• Jianxun Lian
• Jose Hernandez-Orallo
• Aylin Caliskan
• Jindong Wang

Large language models (LLMs) typically generate identical or similar responses for all users given the same prompt, posing serious safety risks in high-stakes applications where user vulnerabilities differ widely. Existing safety evaluations primarily rely on context-independent metrics—such as factuality, bias, or toxicity—overlooking the fact that the same response may carry divergent risks depending on the user's background or condition. We introduce ``personalized safety'' to fill this gap and present PENGUIN—a benchmark comprising 14, 000 scenarios across seven sensitive domains with both context-rich and context-free variants. Evaluating six leading LLMs, we demonstrate that personalized user information significantly improves safety scores by 43. 2%, confirming the effectiveness of personalization in safety alignment. However, not all context attributes contribute equally to safety enhancement. To address this, we develop RAISE—a training-free, two-stage agent framework that strategically acquires user-specific background. RAISE improves safety scores by up to 31. 6% over six vanilla LLMs, while maintaining a low interaction cost of just 2. 7 user queries on average. Our findings highlight the importance of selective information gathering in safety-critical domains and offer a practical solution for personalizing LLM responses without model retraining. This work establishes a foundation for safety research that adapts to individual user contexts rather than assuming a universal harm standard.

EAAI Journal 2025 Journal Article

Reinforcement learning based on multi agent value distribution for beamforming optimization in cellular networks

• Zhibin Liu
• Jindong Wang

This paper considers multi-input single-output (MISO) homogeneous cellular networks, where multi-antenna base stations (BS) in each cellular network use shared spectrum to transmit data to single-antenna users, and employ downlink beamforming techniques to reduce inter-cell interference and improve network energy efficiency (EE). However, due to the high computational cost and related overhead of collecting global channel state information (CSI), most existing techniques are inefficient and impractical in dynamic wireless environments. In this study, we introduce a distributed reinforcement learning framework based on full-parametric quantile function (DRL-FQF) for downlink beamforming, which utilizes deep reinforcement learning (DRL) to dynamically select the optimal beam direction and power allocation according to the environment. DRL-FQF builds upon reinforcement learning algorithms by dynamically optimizing quantile positions and integrating quantile regression to model the distribution of cumulative returns in reinforcement learning. It employs Wasserstein error and quantile Huber loss for learning and self-adjustment, reducing bias in value distribution estimation, thereby replacing the traditional Q-value approach in reinforcement learning. Additionally, the high sample efficiency of DRL-FQF ensures its scalability even in high-dimensional scenarios with an increasing number of base stations, while prioritized experience replay further accelerates convergence. The proposed beamforming framework is efficient and scalable for high-dimensional scenarios with an increasing number of base stations, while greatly reducing information exchange. Simulation results show that the proposed DRL-FQF algorithm achieves higher performance compared to algorithms of existing methods deep Q network (DQN), deep recurrent Q-network (DRQN), and stochastic schemes.

TIST Journal 2024 Journal Article

A Survey on Evaluation of Large Language Models

• Yupeng Chang
• Xu Wang
• Jindong Wang
• Yuan Wu
• Linyi Yang
• Kaijie Zhu
• Hao Chen
• Xiaoyuan Yi

Large language models (LLMs) are gaining increasing popularity in both academia and industry, owing to their unprecedented performance in various applications. As LLMs continue to play a vital role in both research and daily use, their evaluation becomes increasingly critical, not only at the task level, but also at the society level for better understanding of their potential risks. Over the past years, significant efforts have been made to examine LLMs from various perspectives. This paper presents a comprehensive review of these evaluation methods for LLMs, focusing on three key dimensions: what to evaluate, where to evaluate, and how to evaluate. Firstly, we provide an overview from the perspective of evaluation tasks, encompassing general natural language processing tasks, reasoning, medical usage, ethics, education, natural and social sciences, agent applications, and other areas. Secondly, we answer the ‘where’ and ‘how’ questions by diving into the evaluation methods and benchmarks, which serve as crucial components in assessing the performance of LLMs. Then, we summarize the success and failure cases of LLMs in different tasks. Finally, we shed light on several future challenges that lie ahead in LLMs evaluation. Our aim is to offer invaluable insights to researchers in the realm of LLMs evaluation, thereby aiding the development of more proficient LLMs. Our key point is that evaluation should be treated as an essential discipline to better assist the development of LLMs. We consistently maintain the related open-source materials at: https://github.com/MLGroupJLU/LLM-eval-survey

NeurIPS Conference 2024 Conference Paper

CultureLLM: Incorporating Cultural Differences into Large Language Models

• Cheng Li
• Mengzhuo Chen
• Jindong Wang
• Sunayana Sitaram
• Xing Xie

Large language models (LLMs) have been observed to exhibit bias towards certain cultures due to the predominance of training data obtained from English corpora. Considering that multilingual cultural data is often expensive to procure, existing methodologies address this challenge through prompt engineering or culture-specific pre-training. However, these strategies may neglect the knowledge deficiency of low-resource cultures and necessitate substantial computing resources. In this paper, we propose CultureLLM, a cost-effective solution to integrate cultural differences into LLMs. CultureLLM employs the World Value Survey (WVS) as seed data and generates semantically equivalent training data through the proposed semantic data augmentation. Utilizing only $50$ seed samples from WVS with augmented data, we fine-tune culture-specific LLMs as well as a unified model (CultureLLM-One) for $9$ cultures, encompassing both rich and low-resource languages. Extensive experiments conducted on $60$ culture-related datasets reveal that CultureLLM significantly surpasses various counterparts such as GPT-3. 5 (by $8. 1$\%) and Gemini Pro (by $9. 5$\%), demonstrating performance comparable to or exceeding that of GPT-4. Our human study indicates that the generated samples maintain semantic equivalence to the original samples, offering an effective solution for LLMs augmentation. Code is released at https: //github. com/Scarelette/CultureLLM.

NeurIPS Conference 2024 Conference Paper

CulturePark: Boosting Cross-cultural Understanding in Large Language Models

• Cheng Li
• Damien Teney
• Linyi Yang
• Qingsong Wen
• Xing Xie
• Jindong Wang

Cultural bias is pervasive in many large language models (LLMs), largely due to the deficiency of data representative of different cultures. Typically, cultural datasets and benchmarks are constructed either by extracting subsets of existing datasets or by aggregating from platforms such as Wikipedia and social media. However, these approaches are highly dependent on real-world data and human annotations, making them costly and difficult to scale. Inspired by cognitive theories on social communication, this paper introduces CulturePark, an LLM-powered multi-agent communication framework for cultural data collection. CulturePark simulates cross-cultural human communication with LLM-based agents playing roles in different cultures. It generates high-quality cross-cultural dialogues encapsulating human beliefs, norms, and customs. Using CulturePark, we generated 41, 000 cultural samples to fine-tune eight culture-specific LLMs. We evaluated these models across three downstream tasks: content moderation, cultural alignment, and cultural education. Results show that for content moderation, our GPT-3. 5-based models either match or outperform GPT-4 on $41$ datasets. Regarding cultural alignment, our models surpass GPT-4 on Hofstede's VSM 13 framework. Furthermore, for cultural education of human participants, our models demonstrate superior outcomes in both learning efficacy and user experience compared to GPT-4. CulturePark proves an important step in addressing cultural bias and advancing the democratization of AI, highlighting the critical role of culturally inclusive data in model training. Code is released at https: //github. com/Scarelette/CulturePark.

NeurIPS Conference 2024 Conference Paper

Diff-eRank: A Novel Rank-Based Metric for Evaluating Large Language Models

• Lai Wei
• Zhiquan Tan
• Chenghai Li
• Jindong Wang
• Weiran Huang

Large Language Models (LLMs) have transformed natural language processing and extended their powerful capabilities to multi-modal domains. As LLMs continue to advance, it is crucial to develop diverse and appropriate metrics for their evaluation. In this paper, we introduce a novel rank-based metric, Diff-eRank, grounded in information theory and geometry principles. Diff-eRank assesses LLMs by analyzing their hidden representations, providing a quantitative measure of how efficiently they eliminate redundant information during training. We demonstrate the applicability of Diff-eRank in both single-modal (e. g. , language) and multi-modal settings. For language models, our results show that Diff-eRank increases with model size and correlates well with conventional metrics such as loss and accuracy. In the multi-modal context, we propose an alignment evaluation method based on the eRank, and verify that contemporary multi-modal LLMs exhibit strong alignment performance based on our method. Our code is publicly available at https: //github. com/waltonfuture/Diff-eRank.

NeurIPS Conference 2024 Conference Paper

ERBench: An Entity-Relationship based Automatically Verifiable Hallucination Benchmark for Large Language Models

• Jio Oh
• Soyeon Kim
• Junseok Seo
• Jindong Wang
• Ruochen Xu
• Xing Xie
• Steven E. Whang

Large language models (LLMs) have achieved unprecedented performances in various applications, yet evaluating them is still challenging. Existing benchmarks are either manually constructed or are automatic, but lack the ability to evaluate the thought process of LLMs with arbitrary complexity. We contend that utilizing existing relational databases based on the entity-relationship (ER) model is a promising approach for constructing benchmarks as they contain structured knowledge that can be used to question LLMs. Unlike knowledge graphs, which are also used to evaluate LLMs, relational databases have integrity constraints that can be used to better construct complex in-depth questions and verify answers: (1) functional dependencies can be used to pinpoint critical keywords that an LLM must know to properly answer a given question containing certain attribute values; and (2) foreign key constraints can be used to join relations and construct multi-hop questions, which can be arbitrarily long and used to debug intermediate answers. We thus propose ERBench, which uses these integrity constraints to convert any database into an LLM benchmark. ERBench supports continuous evaluation as databases change, multimodal questions, and various prompt engineering techniques. In our experiments, we construct LLM benchmarks using databases of multiple domains and make an extensive comparison of contemporary LLMs. We show how ERBench can properly evaluate any LLM by not only checking for answer correctness, but also effectively verifying the rationales by looking for the right keywords.

NeurIPS Conference 2024 Conference Paper

Imprecise Label Learning: A Unified Framework for Learning with Various Imprecise Label Configurations

• Hao Chen
• Ankit Shah
• Jindong Wang
• Ran Tao
• Yidong Wang
• Xiang Li
• Xing Xie
• Masashi Sugiyama

Learning with reduced labeling standards, such as noisy label, partial label, and supplementary unlabeled data, which we generically refer to as imprecise label, is a commonplace challenge in machine learning tasks. Previous methods tend to propose specific designs for every emerging imprecise label configuration, which is usually unsustainable when multiple configurations of imprecision coexist. In this paper, we introduce imprecise label learning (ILL), a framework for the unification of learning with various imprecise label configurations. ILL leverages expectation-maximization (EM) for modeling the imprecise label information, treating the precise labels as latent variables. Instead of approximating the correct labels for training, it considers the entire distribution of all possible labeling entailed by the imprecise information. We demonstrate that ILL can seamlessly adapt to partial label learning, semi-supervised learning, noisy label learning, and, more importantly, a mixture of these settings, with closed-form learning objectives derived from the unified EM modeling. Notably, ILL surpasses the existing specified techniques for handling imprecise labels, marking the first practical and unified framework with robust and effective performance across various challenging settings. We hope our work will inspire further research on this topic, unleashing the full potential of ILL in wider scenarios where precise labels are expensive and complicated to obtain.

IJCAI Conference 2024 Conference Paper

NegativePrompt: Leveraging Psychology for Large Language Models Enhancement via Negative Emotional Stimuli

• Xu Wang
• Cheng Li
• Yi Chang
• Jindong Wang
• Yuan Wu

Large Language Models (LLMs) have become integral to a wide spectrum of applications, ranging from traditional computing tasks to advanced artificial intelligence (AI) applications. This widespread adoption has spurred extensive research into LLMs across various disciplines, including the social sciences. Notably, studies have revealed that LLMs possess emotional intelligence, which can be further developed through positive emotional stimuli. This discovery raises an intriguing question: can negative emotions similarly influence LLMs, potentially enhancing their performance? In response to this question, we introduce NegativePrompt, a novel approach underpinned by psychological principles, involving ten specifically designed negative emotional stimuli. We embark on rigorous experimental evaluations of five LLMs including Flan-T5-Large, Vicuna, Llama 2, ChatGPT, and GPT-4, across a set of 45 tasks. The results are revealing: NegativePrompt markedly enhances the performance of LLMs, evidenced by relative improvements of 12. 89% in Instruction Induction tasks and 46. 25% in BIG-Bench tasks. Moreover, we conduct attention visualization experiments to decipher the underlying mechanisms of NegativePrompt's influence. Our research contributes significantly to the understanding of LLMs and emotion interaction, demonstrating the practical efficacy of NegativePrompt as an emotion-driven method and offering novel insights for the enhancement of LLMs in real-world applications. The code is available at https: //github. com/wangxu0820/NegativePrompt.

JMLR Journal 2024 Journal Article

PromptBench: A Unified Library for Evaluation of Large Language Models

• Kaijie Zhu
• Qinlin Zhao
• Hao Chen
• Jindong Wang
• Xing Xie

The evaluation of large language models (LLMs) is crucial to assess their performance and mitigate potential security risks. In this paper, we introduce PromptBench, a unified library to evaluate LLMs. It consists of several key components that can be easily used and extended by researchers: prompt construction, prompt engineering, dataset and model loading, adversarial prompt attack, dynamic evaluation protocols, and analysis tools. PromptBench is designed as an open, general, and flexible codebase for research purpose. It aims to facilitate original study in creating new benchmarks, deploying downstream applications, and designing new evaluation protocols. The code is available at: https://github.com/microsoft/promptbench and will be continuously supported. [abs] [ pdf ][ bib ] [ code ] &copy JMLR 2024. ( edit, beta )

NeurIPS Conference 2024 Conference Paper

Slight Corruption in Pre-training Data Makes Better Diffusion Models

• Hao Chen
• Yujin Han
• Diganta Misra
• Xiang Li
• Kai Hu
• Difan Zou
• Masashi Sugiyama
• Jindong Wang

Diffusion models (DMs) have shown remarkable capabilities in generating realistic high-quality images, audios, and videos. They benefit significantly from extensive pre-training on large-scale datasets, including web-crawled data with paired data and conditions, such as image-text and image-class pairs. Despite rigorous filtering, these pre-training datasets often inevitably contain corrupted pairs where conditions do not accurately describe the data. This paper presents the first comprehensive study on the impact of such corruption in pre-training data of DMs. We synthetically corrupt ImageNet-1K and CC3M to pre-train and evaluate over $50$ conditional DMs. Our empirical findings reveal that various types of slight corruption in pre-training can significantly enhance the quality, diversity, and fidelity of the generated images across different DMs, both during pre-training and downstream adaptation stages. Theoretically, we consider a Gaussian mixture model and prove that slight corruption in the condition leads to higher entropy and a reduced 2-Wasserstein distance to the ground truth of the data distribution generated by the corruptly trained DMs. Inspired by our analysis, we propose a simple method to improve the training of DMs on practical datasets by adding condition embedding perturbations (CEP). CEP significantly improves the performance of various DMs in both pre-training and downstream tasks. We hope that our study provides new insights into understanding the data and pre-training processes of DMs.

NeurIPS Conference 2023 Conference Paper

Distilling Out-of-Distribution Robustness from Vision-Language Foundation Models

• Andy Zhou
• Jindong Wang
• Yu-Xiong Wang
• Haohan Wang

We propose a conceptually simple and lightweight framework for improving the robustness of vision models through the combination of knowledge distillation and data augmentation. We address the conjecture that larger models do not make for better teachers by showing strong gains in out-of-distribution robustness when distilling from pretrained foundation models. Following this finding, we propose Discrete Adversarial Distillation (DAD), which leverages a robust teacher to generate adversarial examples and a VQGAN to discretize them, creating more informative samples than standard data augmentation techniques. We provide a theoretical framework for the use of a robust teacher in the knowledge distillation with data augmentation setting and demonstrate strong gains in out-of-distribution robustness and clean accuracy across different student architectures. Notably, our method adds minor computational overhead compared to similar techniques and can be easily combined with other data augmentations for further improvements.

TIST Journal 2022 Journal Article

Domain Generalization for Activity Recognition via Adaptive Feature Fusion

• Xin Qin
• Jindong Wang
• Yiqiang Chen
• Wang Lu
• Xinlong Jiang

Human activity recognition requires the efforts to build a generalizable model using the training datasets with the hope to achieve good performance in test datasets. However, in real applications, the training and testing datasets may have totally different distributions due to various reasons such as different body shapes, acting styles, and habits, damaging the model’s generalization performance. While such a distribution gap can be reduced by existing domain adaptation approaches, they typically assume that the test data can be accessed in the training stage, which is not realistic. In this article, we consider a more practical and challenging scenario: domain-generalized activity recognition (DGAR) where the test dataset cannot be accessed during training. To this end, we propose Adaptive Feature Fusion for Activity Recognition (AFFAR), a domain generalization approach that learns to fuse the domain-invariant and domain-specific representations to improve the model’s generalization performance. AFFAR takes the best of both worlds where domain-invariant representations enhance the transferability across domains and domain-specific representations leverage the model discrimination power from each domain. Extensive experiments on three public HAR datasets show its effectiveness. Furthermore, we apply AFFAR to a real application, i.e., the diagnosis of Children’s Attention Deficit Hyperactivity Disorder (ADHD), which also demonstrates the superiority of our approach.

TMLR Journal 2022 Journal Article

Domain-invariant Feature Exploration for Domain Generalization

• Wang Lu
• Jindong Wang
• Haoliang Li
• Yiqiang Chen
• Xing Xie

Deep learning has achieved great success in the past few years. However, the performance of deep learning is likely to impede in face of non-IID situations. Domain generalization (DG) enables a model to generalize to an unseen test distribution, i.e., to learn domain-invariant representations. In this paper, we argue that domain-invariant features should be originating from both internal and mutual sides. Internal invariance means that the features can be learned with a single domain and the features capture intrinsic semantics of data, i.e., the property within a domain, which is agnostic to other domains. Mutual invariance means that the features can be learned with multiple domains (cross-domain) and the features contain common information, i.e., the transferable features w.r.t. other domains. We then propose DIFEX for Domain-Invariant Feature EXploration. DIFEX employs a knowledge distillation framework to capture the high-level Fourier phase as the internally-invariant features and learn cross-domain correlation alignment as the mutually-invariant features. We further design an exploration loss to increase the feature diversity for better generalization. Extensive experiments on both time-series and visual benchmarks demonstrate that the proposed DIFEX achieves state-of-the-art performance.

NeurIPS Conference 2022 Conference Paper

USB: A Unified Semi-supervised Learning Benchmark for Classification

• Yidong Wang
• Hao Chen
• Yue Fan
• Wang Sun
• Ran Tao
• Wenxin Hou
• Renjie Wang
• Linyi Yang

Semi-supervised learning (SSL) improves model generalization by leveraging massive unlabeled data to augment limited labeled samples. However, currently, popular SSL evaluation protocols are often constrained to computer vision (CV) tasks. In addition, previous work typically trains deep neural networks from scratch, which is time-consuming and environmentally unfriendly. To address the above issues, we construct a Unified SSL Benchmark (USB) for classification by selecting 15 diverse, challenging, and comprehensive tasks from CV, natural language processing (NLP), and audio processing (Audio), on which we systematically evaluate the dominant SSL methods, and also open-source a modular and extensible codebase for fair evaluation of these SSL methods. We further provide the pre-trained versions of the state-of-the-art neural models for CV tasks to make the cost affordable for further tuning. USB enables the evaluation of a single SSL algorithm on more tasks from multiple domains but with less cost. Specifically, on a single NVIDIA V100, only 39 GPU days are required to evaluate FixMatch on 15 tasks in USB while 335 GPU days (279 GPU days on 4 CV datasets except for ImageNet) are needed on 5 CV tasks with TorchSSL.

NeurIPS Conference 2021 Conference Paper

FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labeling

• Bowen Zhang
• Yidong Wang
• Wenxin Hou
• Hao Wu
• Jindong Wang
• Manabu Okumura
• Takahiro Shinozaki

The recently proposed FixMatch achieved state-of-the-art results on most semi-supervised learning (SSL) benchmarks. However, like other modern SSL algorithms, FixMatch uses a pre-defined constant threshold for all classes to select unlabeled data that contribute to the training, thus failing to consider different learning status and learning difficulties of different classes. To address this issue, we propose Curriculum Pseudo Labeling (CPL), a curriculum learning approach to leverage unlabeled data according to the model's learning status. The core of CPL is to flexibly adjust thresholds for different classes at each time step to let pass informative unlabeled data and their pseudo labels. CPL does not introduce additional parameters or computations (forward or backward propagation). We apply CPL to FixMatch and call our improved algorithm FlexMatch. FlexMatch achieves state-of-the-art performance on a variety of SSL benchmarks, with especially strong performances when the labeled data are extremely limited or when the task is challenging. For example, FlexMatch achieves 13. 96% and 18. 96% error rate reduction over FixMatch on CIFAR-100 and STL-10 datasets respectively, when there are only 4 labels per class. CPL also significantly boosts the convergence speed, e. g. , FlexMatch can use only 1/5 training time of FixMatch to achieve even better performance. Furthermore, we show that CPL can be easily adapted to other SSL algorithms and remarkably improve their performances. We open-source our code at https: //github. com/TorchSSL/TorchSSL.

IJCAI Conference 2021 Conference Paper

Generalizing to Unseen Domains: A Survey on Domain Generalization

• Jindong Wang
• Cuiling Lan
• Chang Liu
• Yidong Ouyang
• Tao Qin

Domain generalization (DG), i. e. , out-of-distribution generalization, has attracted increased interests in recent years. Domain generalization deals with a challenging setting where one or several different but related domain(s) are given, and the goal is to learn a model that can generalize to an unseen test domain. For years, great progress has been achieved. This paper presents the first review for recent advances in domain generalization. First, we provide a formal definition of domain generalization and discuss several related fields. Then, we categorize recent algorithms into three classes and present them in detail: data manipulation, representation learning, and learning strategy, each of which contains several popular algorithms. Third, we introduce the commonly used datasets and applications. Finally, we summarize existing literature and present some potential research topics for the future.

NeurIPS Conference 2021 Conference Paper

Learning Causal Semantic Representation for Out-of-Distribution Prediction

• Chang Liu
• Xinwei Sun
• Jindong Wang
• Haoyue Tang
• Tao Li
• Tao Qin
• Wei Chen
• Tie-Yan Liu

Conventional supervised learning methods, especially deep ones, are found to be sensitive to out-of-distribution (OOD) examples, largely because the learned representation mixes the semantic factor with the variation factor due to their domain-specific correlation, while only the semantic factor causes the output. To address the problem, we propose a Causal Semantic Generative model (CSG) based on a causal reasoning so that the two factors are modeled separately, and develop methods for OOD prediction from a single training domain, which is common and challenging. The methods are based on the causal invariance principle, with a novel design in variational Bayes for both efficient learning and easy prediction. Theoretically, we prove that under certain conditions, CSG can identify the semantic factor by fitting training data, and this semantic-identification guarantees the boundedness of OOD generalization error and the success of adaptation. Empirical study shows improved OOD performance over prevailing baselines.

IS Journal 2020 Journal Article

FedHealth: A Federated Transfer Learning Framework for Wearable Healthcare

• Yiqiang Chen
• Xin Qin
• Jindong Wang
• Chaohui Yu
• Wen Gao

With the rapid development of computing technology, wearable devices make it easy to get access to people's health information. Smart healthcare achieves great success by training machine learning models on a large quantity of user personal data. However, there are two critical challenges. First, user data often exist in the form of isolated islands, making it difficult to perform aggregation without compromising privacy security. Second, the models trained on the cloud fail on personalization. In this article, we propose FedHealth, the first federated transfer learning framework for wearable healthcare to tackle these challenges. FedHealth performs data aggregation through federated learning, and then builds relatively personalized models by transfer learning. Wearable activity recognition experiments and real Parkinson's disease auxiliary diagnosis application have evaluated that FedHealth is able to achieve accurate and personalized healthcare without compromising privacy and security. FedHealth is general and extensible in many healthcare applications.

IJCAI Conference 2020 Conference Paper

Joint Partial Optimal Transport for Open Set Domain Adaptation

• Renjun Xu
• Pelen Liu
• Yin Zhang
• Fang Cai
• Jindong Wang
• Shuoying Liang
• Heting Ying
• Jianwei Yin

Domain adaptation (DA) has achieved a resounding success to learn a good classifier by leveraging labeled data from a source domain to adapt to an unlabeled target domain. However, in a general setting when the target domain contains classes that are never observed in the source domain, namely in Open Set Domain Adaptation (OSDA), existing DA methods failed to work because of the interference of the extra unknown classes. This is a much more challenging problem, since it can easily result in negative transfer due to the mismatch between the unknown and known classes. Existing researches are susceptible to misclassification when target domain unknown samples in the feature space distributed near the decision boundary learned from the labeled source domain. To overcome this, we propose Joint Partial Optimal Transport (JPOT), fully utilizing information of not only the labeled source domain but also the discriminative representation of unknown class in the target domain. The proposed joint discriminative prototypical compactness loss can not only achieve intra-class compactness and inter-class separability, but also estimate the mean and variance of the unknown class through backpropagation, which remains intractable for previous methods due to the blindness about the structure of the unknown classes. To our best knowledge, this is the first optimal transport model for OSDA. Extensive experiments demonstrate that our proposed model can significantly boost the performance of open set domain adaptation on standard DA datasets.

TIST Journal 2020 Journal Article

Transfer Learning with Dynamic Distribution Adaptation

• Jindong Wang
• Yiqiang Chen
• Wenjie Feng
• Han Yu
• Meiyu Huang
• Qiang Yang

Transfer learning aims to learn robust classifiers for the target domain by leveraging knowledge from a source domain. Since the source and the target domains are usually from different distributions, existing methods mainly focus on adapting the cross-domain marginal or conditional distributions. However, in real applications, the marginal and conditional distributions usually have different contributions to the domain discrepancy. Existing methods fail to quantitatively evaluate the different importance of these two distributions, which will result in unsatisfactory transfer performance. In this article, we propose a novel concept called Dynamic Distribution Adaptation (DDA), which is capable of quantitatively evaluating the relative importance of each distribution. DDA can be easily incorporated into the framework of structural risk minimization to solve transfer learning problems. On the basis of DDA, we propose two novel learning algorithms: (1) Manifold Dynamic Distribution Adaptation (MDDA) for traditional transfer learning, and (2) Dynamic Distribution Adaptation Network (DDAN) for deep transfer learning. Extensive experiments demonstrate that MDDA and DDAN significantly improve the transfer learning performance and set up a strong baseline over the latest deep and adversarial methods on digits recognition, sentiment analysis, and image classification. More importantly, it is shown that marginal and conditional distributions have different contributions to the domain divergence, and our DDA is able to provide good quantitative evaluation of their relative importance, which leads to better performance. We believe this observation can be helpful for future research in transfer learning.