韩文娟

博士、副教授、副教授

基本信息

办公电话： 电子邮件： wjhan@bjtu.edu.cn
通讯地址：北京交通大学计算机与信息技术学院 邮编：100044

学术主页

中国科学院大学博士，加州大学洛杉矶分校访问学者，新加坡国立大学(NUS)研究员。EMNLP 2022研讨会（UM-IoS Workshop）Chair，ACM SIGAI CHINA 新星奖，铁科院审稿顾问，北京通用人工智能研究院专家顾问。在国内外公认的本学科权威学术刊物上发表40余篇。研究方向为多模态大模型，及在综合交通等安全关键领域的应用。当前的研究重点在于赋予机器基于多模态大模型的认知智能：以多模态联合信息表达机制和跨模态计算建模为切入点和突破口，将高阶常识和认知以符号等形式与低维感知结合，以此增强智能体对人、物体、场景的统一语义表征和抽象推理，构建认知智能增强的交互式智能体。


关于保研、统招和非全硕士生：

  • 实验室团队隶属计算机学院，主要面向（1）层次性自然语言理解，特别是无监督句法分析；（2）多模态时序知识图谱；（3）视觉语言联合理解和解析，构建跨模态统一语义表示进行基础理论研究，主要研究成果是发表高水平论文。
  • 团队为优秀学生提供参加国际会议和联合培养等交流机会，有机会推荐到北京大学、清华大学、微信等机构学习访问。
  • 良好的实验室氛围，团结卓越的团队文化，定期组织篮球、聚餐、郊游活动等。
  • 欢迎计算机基础较好，有程序设计竞赛或者科研经历，有志于攻读硕士研究生和出国深造的同学联系。



实验室网页 Website： Cognition and Language Lab (认知与语言计算实验室*维护中) 

完整论文列表：https://scholar.google.com/citations?user=rfVLLfAAAAAJ&hl=en

实验室 GitHub：https://github.com/cocacola-lab

 

部分项目展示 Selected Project:

ChatIE：通过与ChatGPT对话实现零样本信息抽取：

Demo:   http://124.221.16.143:5000/

GitHub：https://github.com/cocacola-lab/ChatIE

Paper：https://arxiv.org/pdf/2302.10205.pdf


GPT4IE: GPT3.5实现零样本信息抽取：

Demo: http://124.221.16.143:8080/

GitHub：https://github.com/cocacola-lab/GPT4IE


全部项目展示：  https://winniehan.github.io/


教育背景

2014.09-2020.01，中国，中国科学院大学，信息科学与技术学院，通信与信息系统专业，自然语言处理方向，博士，

2019.05-2019.11，美国，加利福尼亚大学洛杉矶分校（University of California, Los Angeles），人工智能方向，访问学者

工作经历

2021/04-2022/10，北京通用人工智能研究院，研究员；

2020/01-2021/04，新加坡国立大学 (National University of Singapore)，计算机系，Research Fellow 博后；

2019/05-2019/11，加利福尼亚大学洛杉矶分校（University of California, Los Angeles），统计系，Visiting Scholar 访问学者

科研项目

  1. 自然科学横向项目: 面向多对多翻译的前沿研究与应用, 2022-2023
  2. 自然科学纵向项目: 国家自然科学基金委员会, 面上项目, 61976139, 低标注资源下的自然语言结构学习, 2020-2023 参与

教学工作

计算语言学，2023年春季，博士生课程

论文/期刊

代表性论著、论文情况(完整列表请关注Google Scholar: https://scholar.google.com/citations?user=rfVLLfAAAAAJ&hl=en)

  • · Z Gao, Y Du, X Zhang, X Ma, Wenjuan Han, SC Zhu, Q Li. CLOVA: A Closed-Loop Visual Assistant with Tool Usage and Update. (CVPR 2024)

    · Haozhe Zhao, Zefan Cai, Shuzheng Si, Xiaojian Ma, Kaikai An, Liang Chen, Zixuan Liu, Sheng Wang, Wenjuan Han, Baobao Chang, Mmicl: Empowering vision-language model with multi-modal in-context learning. (ICLR 2024)

    · CollabKG: A Learnable Human-Machine-Cooperative Information Extraction Toolkit for (Event) Knowledge Graph Construction. (COLING2024)

    · Empowering Vision-Language Models for Reasoning Ability through Large Language Models. (ICASSP2024)

    · Zhe,Manjie,Hopcroft,Kun,Tenenbaum,Chun,Nian,韩文娟,Yixin.On the Complexity of Bayesian Generalization.Proceedings of the 40th International Conference on Machine Learning (ICML2023),

    · Wenjuan Han ,Zhao, Cai.Empowering MultiModal Models’ In-Context Learning Ability through Large Language Models. Proceedings of the ACM Turing Award Celebration Conference,2023

    · Xue Zhang,Songming Zhang,Yunlong Liang, Yufeng Liang, Jian Liu, Wenjuan Han, Jinan Xu. A Quality-based Syntactic Template Retriever for Syntactically-Controlled Paraphrase Generation.2023 Conference on Empirical Methods in Natural Language Processing, (ACL2023)

    · Jiapeng Li, Wenjuan Han, Lifeng Fan. IntentQA: Context-aware Video Intent Reasoning.International Conference on Computer Vision (ICCV2023)

    · Guangyuan Jiang, Manjie Xu, Wenjuan Han, Chi Zhang, Yixin Zhu. Evaluating and Inducing Personality in Pre-trained Language Models.37th Conference on Neural Information Processing Systems (NeurIPS 2023)

    · Li,Hu,Wenjuan Han,Zou. CORD: A Three-Stage Coarse-to-Fine Framework for Relation Detection in Knowledge Base Question Answering.the 32nd ACM International Conference on Information and Knowledge Management,(CIKM2023)

    · Jia,Yan,Wenjuan Han, Zheng,Tu.Modeling Instance Interactions for Joint Information Extraction with Neural High-Order Conditional Random Field.61st Annual Meeting of the Association for Computational Linguistics,(ACL2023)

    · Hanmin Wu,Wenjuan Han,Yufeng Chen. Jinan Xu. A Holistic Approach to Reference-Free Evaluation of Machine Translation.The 61st Annual Meeting of the Association for Computational Linguistics, (ACL2023)

    · Yuzhe Shi, Wenjuan Han. et al. PersLEARN: Research Training through the Lens of Perspective Cultivation.The 61st Annual Meeting of the Association for Computational Linguistics, (ACL2023)

    · Songmin Zhang,Wenjuan Han, et al. Towards Understanding and Improving Knowledge Distillation for Neural Machine Translation.The 61st Annual Meeting of the Association for Computational Linguistics （ACL 2023）

    · Chao Lou, Wenjuan Han (通讯), Yuhuan Lin, Zilong Zheng. Unsupervised Visual-Language Parsing: Seamlessly Bridging Visual Scene Graphs with Language Structures via Dependency Relationships. CVPR 2022

    · Wan Bo, Wenjuan Han (通讯), Zilong Zheng, Tinne Tuytelaars. Unsupervised Vision-Language Grammar Induction with Shared Structure Modeling (Oral, Score: 8). ICLR 2022

    · Ye Hai, Hwee Ng, Wenjuan Han. On the Robustness of Question Rewriting Systems to Questions of Varying Hardness. ACL 2022 

    · Wenjuan Han, Hwee Tou Ng. Diversity-Driven Combination for Grammatical Error Correction. ICTAI 2021. 

    · Wenjuan Han (通讯), Bo Pang, Yingnian. Robust Transfer Learning with Pretrained Language Models through Adapters. ACL 2021

    · Bo Pang, Erik Nijkamp, Wenjuan Han（通讯）, Linqi Zhou, Yixian Liu and Kewei Tu. Towards Holistic and Automatic Evaluation of Open-Domain Dialogue Generation. ACL 2020 

    · Wenjuan Han, Jiang Yong, Hwee Tou Ng and Kewei Tu. A Survey of Unsupervised Dependency Parsing. COLING 2020

    · Wenjuan Han, Liwen Zhang, Jiang Yong and Kewei Tu. Adversarial Attack and Defense of Structured Prediction Models. EMNLP 2020 

    · Wenjuan Han, Yong Jiang, Kewei Tu. Enhancing Unsupervised Generative Dependency Parser with Contextual Information. ACL 2019

    · Wenjuan Han, Ge Wang, Yong Jiang, Kewei Tu. Multilingual Grammar Induction with Continuous Language Identification. EMNLP 2019

    · Wenjuan Han, Yong Jiang, Kewei Tu. Dependency Grammar Induction with Neural Lexicalization and Big Training Data. EMNLP 2017

    · Wenjuan Han, Yong Jiang, Kewei Tu. Lexicalized neural unsupervised dependency parsing. Neurocomputing 2019

    · Yong Jiang, Wenjuan Han, Kewei Tu. A Regularization-based Framework for Bilingual Grammar Induction. EMNLP 2019 

    · Yong Jiang, Wenjuan Han, Kewei Tu. Combining Generative and Discriminative Approaches to Unsupervised Dependency Parsing via Dual Decomposition. EMNLP 2017

    · Yong Jiang, Wenjuan Han, Kewei Tu. Unsupervised Neural Dependency Parsing. EMNLP 2016

    · Wenjuan Han (1st Workshop Organizers), Yoon Kim, Kewei Tu et.al., UM-IoS: Unimodal and Multimodal Induction of Linguistic Structures. EMNLP 2022

    · Liwen Zhang, Ge Wang, Wenjuan Han, Kewei Tu. Adapting Unsupervised Syntactic Parsing Methodology for Discourse Dependency Parsing. ACL 2021

    · Unsupervised Natural Language Parsing. Kewei Tu, Yong Jiang, Wenjuan Han, Yanpeng Zhao. EACL 2021

    · Second-Order Unsupervised Neural Dependency Parsing. Songlin Yang, Yong Jiang, Wenjuan Han and Kewei Tu. COLING 2020.

    · ToHRE: A Top-Down Classification Strategy with Hierarchical Bag Representation for Distantly Supervised Relation Extraction. Erxin Yu, Wenjuan Han, Yuan Tian and Yi Chang. COLING 2020

    · VGStore: A Multimodal Extension to SPARQL for Querying RDF Scene Graph. Yanzeng Li, Zilong Zheng, Wenjuan Han,  Lei Zou. ISWC 2022

    · Spa: On the Sparsity of Virtual Adversarial Training for Dependency Parsing. Chao Lou, Wenjuan Han, Kewei Tu. AACL 2022

     



专著/译著

《Synthesis Lectures on Human Language Technologies》一部 （In Process）

专利

  1. 韩文娟; 郑子隆 ; 对话系统的评估方法、装置、电子设备及存储介质, 2022-09-19, 中国, CN202211139728.3 
  2. 陈浩; 于雪莉; 刘智静; 王华卿; 韩文娟; 弓琰 ; 一种可解释的智能判决方法、装置、电子设备及存 储介质, 2022-07-29, 中国, CN202210908771.5 
  3. 吴侠宝, 张燕华, 王琦, 韩文娟等, 一种光纤传能系统联锁保护装置及其实现方法. 发表号: CN104009451A.

获奖与荣誉

EMNLP 2022研讨会（UM-IoS Workshop）Chair

ACM SIGAI CHINA 2021新星奖

学术活动

举办国际学术会议：

2022-12-7至2022-12-7

举办Proceedings of the Workshop on Unimodal and Multimodal Induction of Linguistic Structures (UM-IoS) (EMNLP 2022 Workshop)

地点：Abu Dhabi and Virtual

发起人、组织者：韩文娟 

 

特邀报告

受邀分别在南加州大学、欧洲顶级自然语言处理会议EACL作专题讲座

 

国内外学术组织、学术会议重要职务：

审稿人: ACL | TKDE | WWW(TheWebConf) | AAAI | IJCAI | EMNLP | NAACL

长期审稿人: Computational Linguistics

分会主席: COLING2020 (Session Chair)