Xuanjing Huang (黄萱菁)

Xuanjing Huang (黄萱菁)

Professor at Fudan University, 复旦大学教授

Posts by Collection

portfolio

publications

Adversarial Multi-task Learning for Text Classification

Published in Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, 2017

The paper proposed an adversarial multi-task learning framework, alleviating the shared and private latent feature spaces from interfering with each other.

Recommended citation: Pengfei Liu, Xipeng Qiu, Xuanjing Huang: Adversarial Multi-task Learning for Text Classification. ACL (1) 2017: 1-10 http://xuanjing-huang.github.io/files/AMT.pdf

Adversarial Multi-Criteria Learning for Chinese Word Segmentation

Published in Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, 2017

In this paper, we propose adversarial multi-criteria learning for CWS by integrating shared knowledge from multiple heterogeneous segmentation criteria.

Recommended citation: Xinchi Chen, Zhan Shi, Xipeng Qiu, Xuanjing Huang: Adversarial Multi-Criteria Learning for Chinese Word Segmentation. ACL (1) 2017: 1193-1203 http://xuanjing-huang.github.io/files/cws.pdf

A Lexicon-Based Graph Neural Network for Chinese NER

Published in Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), 2019

In this work, we introduce a lexicon-based graph neural network with global semantics for Chinese NER.

Recommended citation: Tao Gui, Yicheng Zou, Qi Zhang, Minlong Peng, Jinlan Fu, Zhongyu Wei, Xuanjing Huang: A Lexicon-Based Graph Neural Network for Chinese NER. EMNLP/IJCNLP (1) 2019: 1040-1050 http://xuanjing-huang.github.io/files/ALB.pdf

FLAT: Chinese NER Using Flat-Lattice Transformer

Published in Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020

In this paper, we propose FLAT: Flat-LAttice Transformer for Chinese NER, which converts the lattice structure into a flat structure consisting of spans.

Recommended citation: Xiaonan Li, Hang Yan, Xipeng Qiu, Xuanjing Huang: FLAT: Chinese NER Using Flat-Lattice Transformer. ACL 2020: 6836-6842 http://xuanjing-huang.github.io/files/FLAT.pdf

Simplify the Usage of Lexicon in Chinese NER

Published in Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020

In this work, we propose a simple but effective method for incorporating the word lexicon into the character representations.

Recommended citation: Ruotian Ma, Minlong Peng, Qi Zhang, Zhongyu Wei, Xuanjing Huang: Simplify the Usage of Lexicon in Chinese NER. ACL 2020: 5951-5960 http://xuanjing-huang.github.io/files/Simplify.pdf

Extractive Summarization as Text Matching

Published in Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020

This paper creates a paradigm shift with regard to the way we build neural extractive summarization systems.

Recommended citation: Ming Zhong, Pengfei Liu, Yiran Chen, Danqing Wang, Xipeng Qiu, Xuanjing Huang: Extractive Summarization as Text Matching. ACL 2020: 6197-6208 http://xuanjing-huang.github.io/files/ext.pdf

K-Adapter: Infusing Knowledge into Pre-Trained Models with Adapters

Published in Findings of the Association for Computational Linguistics: ACL-IJCNLP, 2021

The paper proposes a framework that retains the original parameters of the pre-trained model fixed and supports the development of versatile knowledge-infused model.

Recommended citation: Ruize Wang, Duyu Tang, Nan Duan, Zhongyu Wei, Xuanjing Huang, Jianshu Ji, Guihong Cao, Daxin Jiang, Ming Zhou: K-Adapter: Infusing Knowledge into Pre-Trained Models with Adapters. ACL/IJCNLP (Findings) 2021: 1405-1418 http://xuanjing-huang.github.io/files/K-Adapter.pdf

A Multi-Format Transfer Learning Model for Event Argument Extraction via Variational Information Bottleneck

Published in Proceedings of the 29th International Conference on Computational Linguistics, 2022

In this paper, we propose a multi-format transfer learning model with variational information bottleneck for EAE in new datasets.

Recommended citation: Jie Zhou, Qi Zhang, Qin Chen, Liang He, Xuanjing Huang: A Multi-Format Transfer Learning Model for Event Argument Extraction via Variational Information Bottleneck. COLING 2022: 1990-2000 http://xuanjing-huang.github.io/files/mft.pdf

Secrets of RLHF in Large Language Models Part I: PPO

Published in CoRR abs/2307.04964, 2023

We dissect the framework of RLHF, re-evaluate the inner workings of PPO, and explore how the parts comprising PPO algorithms impact policy agent training.

Recommended citation: Rui Zheng, Shihan Dou, Songyang Gao, Yuan Hua, Wei Shen, Binghai Wang, Yan Liu, Senjie Jin, Qin Liu, Yuhao Zhou, Limao Xiong, Lu Chen, Zhiheng Xi, Nuo Xu, Wenbin Lai, Minghao Zhu, Cheng Chang, Zhangyue Yin, Rongxiang Weng, Wensen Cheng, Haoran Huang, Tianxiang Sun, Hang Yan, Tao Gui, Qi Zhang, Xipeng Qiu, Xuanjing Huang: Secrets of RLHF in Large Language Models Part I: PPO. CoRR abs/2307.04964 (2023) http://xuanjing-huang.github.io/files/rlhf.pdf

Introduction to Natural Language Processing

Published in Electronic Industry Press, 2023

With the widespread application of natural language processing and the rapid advancement of machine learning algorithms represented by deep learning, natural language processing algorithms and research tasks have been developing rapidly in recent years. Since 2003, the authors have taught natural language processing courses for undergraduates, master students, and doctoral students at the School of Computer Science and Technology, Fudan University. This book summarizes years of teaching and research, aiming to provide readers with a more systematic and comprehensive understanding of natural language processing.

Recommended citation: Qi Zhang, Tao Gui, Xuanjing Huang: Introduction to Natural Language Processing, Electronic Industry Press, 2023 https://intro-nlp.github.io/

自然语言处理导论

Published in 电子工业出版社, 2023

随着自然语言处理的广泛应用以及以深度学习为代表的机器学习算法的快速进步，近年来自然语言处理算法和研究任务也在快速发展中。作者自2003年起，在复旦大学计算机科学技术学院针对本科生、硕士生和博士生先后分别开设了自然语言处理课程。本书对多年教学和研究进行总结梳理，希望使得读者对自然语言处理有更加系统性且全面的了解。

Recommended citation: 张奇、桂韬、黄萱菁：自然语言处理导论，电子工业出版社，2023 https://intro-nlp.github.io/

MouSi: Poly-Visual-Expert Vision-Language Models

Published in CoRR abs/2401.17221, 2024

This paper proposes the use of ensemble experts technique to synergizes the capabilities of individual visual encoders, including those skilled in image-text matching, OCR, image segmentation, etc.

Recommended citation: Xiaoran Fan, Tao Ji, Changhao Jiang, Shuo Li, Senjie Jin, Sirui Song, Junke Wang, Boyang Hong, Lu Chen, Guodong Zheng, Ming Zhang, Caishuang Huang, Rui Zheng, Zhiheng Xi, Yuhao Zhou, Shihan Dou, Junjie Ye, Hang Yan, Tao Gui, Qi Zhang, Xipeng Qiu, Xuanjing Huang, Zuxuan Wu, Yu-Gang Jiang: MouSi: Poly-Visual-Expert Vision-Language Models. CoRR abs/2401.17221 (2024) http://xuanjing-huang.github.io/files/mousi.pdf

Secrets of RLHF in Large Language Models Part II: Reward Modeling

Published in CoRR abs/2401.06080, 2024

From a data perspective, we propose a method to measure the strength of preferences within the data, based on a voting mechanism of multiple reward models. From an algorithmic standpoint, we introduce contrastive learning to enhance the ability of reward models to distinguish between chosen and rejected responses, thereby improving model generalization.

Recommended citation: Binghai Wang, Rui Zheng, Lu Chen, Yan Liu, Shihan Dou, Caishuang Huang, Wei Shen, Senjie Jin, Enyu Zhou, Chenyu Shi, Songyang Gao, Nuo Xu, Yuhao Zhou, Xiaoran Fan, Zhiheng Xi, Jun Zhao, Xiao Wang, Tao Ji, Hang Yan, Lixing Shen, Zhan Chen, Tao Gui, Qi Zhang, Xipeng Qiu, Xuanjing Huang, Zuxuan Wu, Yu-Gang Jiang: Secrets of RLHF in Large Language Models Part II: Reward Modeling. CoRR abs/2401.06080 (2024) http://xuanjing-huang.github.io/files/reward.pdf

Searching for Best Practices in Retrieval-Augmented Generation

Published in Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, 2024

Retrieval-augmented generation (RAG) techniques have proven to be effective in integrating up-to-date information, mitigating hallucinations, and enhancing response quality, particularly in specialized domains. While many RAG approaches have been proposed to enhance large language models through query-dependent retrievals, these approaches still suffer from their complex implementation and prolonged response times.

Recommended citation: Xiaohua Wang, Zhenghua Wang, Xuan Gao, Feiran Zhang, Yixin Wu, Zhibo Xu, Tianyuan Shi, Zhengyuan Wang, Shizheng Li, Qi Qian, Ruicheng Yin, Changze Lv, Xiaoqing Zheng, Xuanjing Huang. Searching for Best Practices in Retrieval-Augmented Generation. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 17716–17736. Association for Computational Linguistics, 2024. https://aclanthology.org/2024.emnlp-main.981/

The Rise and Potential of Large Language Model Based Agents: A Survey

Published in Science China Information Sciences, 2025

In this paper, we perform a comprehensive survey on LLM-based agents.

Recommended citation: Zhiheng Xi, Wenxiang Chen, Xin Guo, Wei He, Yiwen Ding, Boyang Hong, Ming Zhang, Junzhe Wang, Senjie Jin, Enyu Zhou, Rui Zheng, Xiaoran Fan, Xiao Wang, Limao Xiong, Yuhao Zhou, Weiran Wang, Changhao Jiang, Yicheng Zou, Xiangyang Liu, Zhangyue Yin, Shihan Dou, Rongxiang Weng, Wenjuan Qin, Yongyan Zheng, Xipeng Qiu, Xuanjing Huang, Qi Zhang, Tao Gui: The rise and potential of large language model based agents: a survey. Sci. China Inf. Sci. 68(2) (2025) https://link.springer.com/article/10.1007/s11432-024-4222-0

Large Language Models: From Theory to Practice (2nd Edition)

Published in Electronic Industry Press, 2025

This book introduces the fundamental theories of large language models, including language modeling, distributed model training, and reinforcement learning, with practical examples using the Deepspeed-Chat framework to implement large language models and ChatGPT-like systems.

Recommended citation: Qi Zhang, Tao Gui, Rui Zheng, Xuanjing Huang: Large Language Models: From Theory to Practice (2nd Edition), Electronic Industry Press, 2025 https://intro-llm.github.io/

大规模语言模型：从理论与实践（第2版）

Published in 电子工业出版社, 2025

本书将介绍大语言模型的基础理论包括语言模型、分布式模型训练以及强化学习，并以Deepspeed-Chat框架为例介绍实现大语言模型和类ChatGPT系统的实践。

Recommended citation: 张奇、桂韬、郑锐、黄萱菁：大规模语言模型：从理论与实践（第2版），电子工业出版社，2025 https://intro-llm.github.io/

AgentGym: Evaluating and Training Large Language Model-based Agents across Diverse Environments

Published in Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 2025

Large language models (LLMs) have emerged as a promising foundation to build generally-capable agents (LLM-based agents) that can handle multi-turn decision-making tasks across various environments. However, the community lacks a unified interactive framework that covers diverse environments for comprehensive evaluation of agents, and enables exploration and learning for their self-improvement.

Recommended citation: Zhiheng Xi, Yiwen Ding, Wenxiang Chen, Boyang Hong, Honglin Guo, Junzhe Wang, Xin Guo, Dingwen Yang, Chenyang Liao, Wei He, Songyang Gao, Lu Chen, Rui Zheng, Yicheng Zou, Tao Gui, Qi Zhang, Xipeng Qiu, Xuanjing Huang, Zuxuan Wu, Yu-Gang Jiang. AgentGym: Evaluating and Training Large Language Model-based Agents across Diverse Environments. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 27914–27961, Vienna, Austria. Association for Computational Linguistics, 2025. https://aclanthology.org/2025.acl-long.1355/

AgentGym-RL: Training LLM Agents for Long-Horizon Decision Making through Multi-Turn Reinforcement Learning

Published in arXiv, 2025

We propose AgentGym-RL, a reinforcement learning framework for training large language model (LLM)-based agents to tackle long-horizon decision-making tasks through multi-turn interactions.

Recommended citation: Zhiheng Xi, Yiwen Ding, Wenxiang Chen, Boyang Hong, Honglin Guo, Junzhe Wang, Xin Guo, Dingwen Yang, Chenyang Liao, Wei He, Songyang Gao, Lu Chen, Rui Zheng, Yicheng Zou, Tao Gui, Qi Zhang, Xipeng Qiu, Xuanjing Huang, Zuxuan Wu, Yu-Gang Jiang. AgentGym-RL: Training LLM Agents for Long-Horizon Decision Making through Multi-Turn Reinforcement Learning. arXiv:2509.08755, 2025. https://arxiv.org/abs/2509.08755

Revealing emergent human-like conceptual representations from language prediction

Published in Proceedings of the National Academy of Sciences (PNAS), 2025

People acquire concepts through rich physical and social experiences and use them to understand and navigate the world. In contrast, large language models (LLMs), trained solely through next-token prediction on text, exhibit strikingly human-like behaviors. Are these models developing concepts akin to those in humans?

Recommended citation: Ningyu Xu, Qi Zhang, Chenyang Du, Qinan Luo, Xipeng Qiu, Xuanjing Huang, Menghan Zhang. Revealing emergent human-like conceptual representations from language prediction. Proceedings of the National Academy of Sciences (PNAS), 2025. https://doi.org/10.1073/pnas.2512514122

CL-bench: A Benchmark for Context Learning

Published in arXiv, 2026

Current language models (LMs) excel at reasoning over prompts using pre-trained knowledge. However, real-world tasks are far more complex and context-dependent: models must learn from task-specific context and leverage new knowledge beyond what is learned during pre-training to reason and resolve tasks. We term this capability context learning, a crucial ability that humans naturally possess but has been largely overlooked.

Recommended citation: Shihan Dou, Ming Zhang, Zhangyue Yin, Chenhao Huang, Yujiong Shen, Junzhe Wang, Jiayi Chen, Yuchen Ni, Junjie Ye, Cheng Zhang, Huaibing Xie, Jianglu Hu, Shaolei Wang, Weichao Wang, Yanling Xiao, Yiting Liu, Zenan Xu, Zhen Guo, Pluto Zhou, Tao Gui, Zuxuan Wu, Xipeng Qiu, Qi Zhang, Xuanjing Huang, Yu-Gang Jiang, Di Wang, Shunyu Yao. CL-bench: A Benchmark for Context Learning. arXiv:2602.03587, 2026. https://arxiv.org/abs/2602.03587

talks

Intelligent Social Media Mining Based on Deep Learning

Published:

Social media refers to various media platforms through which people communicate in online society, possessing significant commercial and social value, and serving as important channels for information dissemination and maintaining social relationships. Over the past few years, the Natural Language Processing team at Fudan University has conducted various intelligent mining research on social media, forming a chain of social media understanding, discovery, and prediction. This includes understanding non-standard textual content on social media, discovering valuable information from social media, and predicting user behavior on social media. This talk mainly introduces methods for predicting user behavior on social media, including microblog tag recommendation, @user (company) recommendation, retweet behavior prediction, user topic participation prediction, expert recommendation, and incorporating multimodal information in social media mining.

基于深度学习的智能社会媒体挖掘

Published:

社交媒体是对人们在网络社会进行沟通的各种媒体的总称，具有重要的商业价值和社会价值，也是信息传播和维系社会关系的重要渠道。过去几年，复旦大学的自然语言处理团队在社会媒体开展各种智能挖掘研究，形成了社会媒体理解、发现、预测的链条，包括理解社会媒体上非规范的文字内容，从社会媒体发现有价值的信息，预测社会媒体上的用户行为。这个报告了主要介绍社交媒体上的用户行为预测方法，包括微博标签推荐、@用户(公司)推荐、转发行为预测、用户话题参与预测、专家推荐、在社会媒体挖掘中融入多模态信息等。

Robustness Issues in Information Extraction

Published:

Information extraction mainly includes two major tasks: named entity recognition and relation extraction, aiming to automatically extract key information from massive unstructured text, thereby effectively supporting downstream tasks such as knowledge graph construction and intelligent question answering. In the era of deep learning, since neural networks, particularly pre-trained models, can automatically extract high-level semantic features, more attention has been focused on how to construct pre-training tasks to achieve more comprehensive semantic knowledge embedding and how to efficiently use such models. However, the automatic feature extraction of deep learning models inevitably leads to shortcut learning problems, resulting in robustness deficiencies in real-world application scenarios, posing hidden dangers for downstream applications of information extraction, especially severe in low-resource environments. This talk will conduct an in-depth analysis of robustness issues in information extraction, explore the underlying reasons affecting model robustness, and introduce our research achievements in improving the robustness of information extraction models in scenarios such as weak samples, few-shot learning, unlabeled data, and cross-domain settings.

信息提取的鲁棒性问题

Published:

信息提取主要包括命名实体识别及关系提取两大主要任务，旨在自动地从海量非结构化文本中抽取出关键信息，从而有效地支撑知识图谱构建和智能问答等下游任务。在深度学习时代，由于神经网络，特别是预训练模型已经能自动地提取高层语义特征，人们把更多的精力关注在如何构建预训练任务实现更完备的语义知识嵌入，以及如何高效使用这样的模型。然而，深度学习模型自动提取特征难以避免捷径学习问题，导致现实应用场景下的鲁棒性缺陷，对信息提取的下游应用带来了一些隐藏的危险，在低资源环境下尤为严重。本报告将围绕信息提取的鲁棒性问题展开深入分析，探究影响模型鲁棒性的深层原因，并介绍我们在弱样本、小样本、无标注、跨领域等场景上提升信息提取模型鲁棒性的研究成果。

Incorporating Dynamic Structures into Pre-trained Language Models

Published:

Recent years have witnessed the great success of large-scale pre-trained language models. However, performing the entire language model for each sample can be computationally uneconomical. Hence, dynamic networks are attracting a lot of attention in the NLP community, which can adapt their structures or parameters to the input samples during inference. In contrast to static language models, dynamic ones enjoy favorable properties such as efficiency, adaptiveness, accuracy, etc. In this talk, I will review recent advances in dynamic networks in NLP and discuss the prospects and challenges of applying dynamic structure to pre-trained language models.

Interpretability Analysis of Artificial Intelligence - A Case Study in Natural Language Processing

Published:

Interpretability in machine learning and deep learning refers to the extent to which model predictions can be explained in a way that is understandable and straightforward to the audience. In recent years, deep learning has achieved successful applications in natural language processing, significantly improving the performance of various tasks. However, due to its inherent complexity, the comprehensibility and interpretability are not satisfactory, which also hinders the further promotion of deep learning methods. This talk first introduces what interpretability analysis is, what interpretability analysis tasks exist in natural language processing, and the purpose of interpretability analysis. Then, it introduces the current development status of interpretability analysis in the field of natural language processing from three aspects: understanding the functional attributes of model components, explaining the behavior of model predictions, and model diagnosis. Finally, it discusses future research trends.

人工智能的的可解释性分析–以自然语言处理为例

Published:

机器学习和深度学习的可解释性指的是以受众可理解的，直截了当的方式解释模型预测值的程度。近年来，深度学习已经在自然语言处理中取得成功应用，大幅度提升了各种任务的性能，但由于其内在复杂性，可理解性和可解释性不够令人满意，也妨碍了深度学习方法的进一步推广。该报告首先介绍什么是可解释性分析，自然语言处理中有哪些可解释性分析任务，可解释性分析的目的，然后从理解模型部件的功能属性、解释模型预测的行为、模型诊断三个方面介绍可解释性分析在自然语言处理领域的发展现状，最后讨论了未来的研究趋势。

Natural Language Processing and Representation Learning

Published:

Natural language typically refers to human language, which serves as a carrier of logical thinking, a means of communication, and a vehicle for cultural heritage. The processing of natural language is an important research content in artificial intelligence, often referred to as the pearl on the crown of AI. An essential foundational step in natural language processing is language representation learning, which aims to construct formal or mathematical descriptions of natural language so that it can be represented in computers and automatically processed by computer programs. Early language representation methods mainly used symbolic discrete representations. In recent years, deep neural networks have been widely applied in natural language processing, not only achieving performance surpassing traditional statistical methods in many tasks such as text classification, sequence labeling, machine translation, and automatic question answering, but also enabling end-to-end training, avoiding cumbersome feature engineering. The first part of this talk will introduce the basic tasks, application areas, research history, and technological development trends of natural language processing. The second part will introduce neural network-based language representation learning methods at various granularities such as words, phrases, sentences, and sentence pairs, explaining how the latent grammatical or semantic features of language are distributedly stored in a group of neurons and represented using dense, low-dimensional, continuous vectors. It will also discuss recent research trends in neural language representation learning from the perspectives of models and learning.

自然语言处理与表示学习

Published:

自然语言通常指人类的语言，是思维逻辑的载体，交流沟通的方式，也是传承文明的手段。对自然语言的处理是人工智能的重要研究内容，被称为人工智能皇冠上的明珠。自然语言处理必不可少的基础步骤是语言表示学习，其目的是构建自然语言的形式化或数学描述，以便在计算机中表示自然语言，并能让计算机程序进行自动处理。早期的语言表示方法主要采用符号化的离散表示。近年来，深度神经网络广泛应用于自然语言处理，不仅在文本分类、序列标注、机器翻译和自动问答等许多任务中取得了超越传统统计方法的性能，而且能以端到端的方式进行训练，避免了繁琐的特征工程。报告的第一部分将介绍自然语言处理的基本任务、应用领域、研究历史和技术发展趋势；第二部分将从词语、短语、句子和句对等粒度介绍基于神经网络的语言表示学习方法，阐述如何将语言的潜在语法或语义特征分布式地存储在一组神经元中，用稠密、低维、连续的向量来表示，并从模型、学习等层面讨论神经语言表示学习的近期研究趋势。

teaching