1 课题组介绍/导师介绍

蒋智威，必威西汉姆联官网助理教授（特聘研究员），博士生导师。2018年6月于必威betway西汉姆联网站获得博士学位。主要研究方向为自然语言处理和多模态学习，包括大语言模型、文本理解与表示学习、音视频理解与生成等。此外，还关注自然语言处理与其他相关领域的交叉研究，如软工文本挖掘、软件缺陷定位等。近年来在ACL、NeurIPS、ICLR、SIGIR、WWW、AAAI、IJCAI、MM、TOIS等国际一流期刊和会议上发表学术论文40余篇。已指导学生获得多项荣誉，包括中国科协青年人才托举工程-博士生专项，博士生国家奖学金，必威betway西汉姆联网站优秀博士论文提名，腾讯犀牛鸟精英人才计划等。

2 课题组研究方向

课题组主要围绕大语言模型、自然语言理解、多模态学习三个大方向来开展前沿探索。

(1)大语言模型方向：

在大语言模型方向，主要关注模型的可控生成、模型幻觉、模型安全以及垂域应用等方面的研究。代表性工作包括：

●可控生成：基于模型干预的句子嵌入生成

●模型幻觉：基于回溯和向量引导的幻觉检测与移除

●模型安全：基于语义水印的模型版权保护

●垂域应用：基于大语言模型精排序的缺陷定位

(2)自然语言理解方向：

在自然语言理解方向，主要关注文本内容分析和文本质量评估。

●文本内容分析：文本分类、主题分析、情感分析

●文本质量评估：文本排序、可读性评估、作文评分

(3)多模态学习方向：

在多模态学习方向，主要关注多模态理解以及多模态生成。代表性工作包括：

●多模态理解：音视频理解

●多模态生成：图像生成、肖像视频生成、手语视频生成

3 课题组成果

课题组近五年来在ACL、NeurIPS、ICLR、SIGIR、WWW、AAAI、IJCAI、MM、TOIS等国际一流期刊和会议上发表学术论文40余篇。

近三年的代表性学术成果：

[1]Yuchen Fu, Zifeng Cheng, Zhiwei Jiang*, Zhonghui Wang, Yafeng Yin, Zhengliang Li, Qing Gu. Token Prepending: A Training-Free Approach for Eliciting Better Sentence Embeddings from LLMs. In Proceedings of the 63st Annual Meeting of the Association for Computational Linguistics (ACL). 2025. (CCF-A)

[2]Zifeng Cheng, Zhonghui Wang, Yuchen Fu, Zhiwei Jiang*, Yafeng Yin, Cong Wang, Qing Gu. Contrastive Prompting Enhances Sentence Embeddings in LLMs through Inference-Time Steering. In Proceedings of the 63st Annual Meeting of the Association for Computational Linguistics (ACL). 2025. (CCF-A)

[3]Shiping Ge, Qiang Chen, Zhiwei Jiang*, Yafeng Yin, Liu Qin, Ziyao Chen, and Qing Gu. Implicit Location-Caption Alignment via Complementary Masking for Weakly-Supervised Dense Video Captioning. In Proceedings of AAAI Conference on Artificial Intelligence (AAAI), 2025. (CCF-A)

[4]Yuchen Fu, Zhiwei Jiang*, Yuliang Liu, Cong Wang, Zexuan Deng, Zhaoling Chen, and Qing Gu. AP-Adapter: Improving Generalization of Automatic Prompts on Unseen Text-to-Image Diffusion Models. In Proceedings of the 38th Annual Conference on Neural Information Processing Systems (NeurIPS). 2024. (CCF-A)

[5]Shiping Ge, Qiang Chen, Zhiwei Jiang*, Yafeng Yin, Ziyao Chen, and Qing Gu. Short Video Ordering via Position Decoding and Successor Prediction. In Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR). 2024: 2167-2176. (CCF-A)

[6]Zhiwei Jiang*, Tianyi Gao, Yafeng Yin, Meng Liu, Hua Yu, Zifeng Cheng, and Qing Gu. Improving Domain Generalization for Prompt-Aware Essay Scoring via Disentangled Representation Learning. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL). 2023: 12456-12470. (CCF-A)

[7]Cong Wang, Zhiwei Jiang*, Yafeng Yin, Zifeng Cheng, Shiping Ge, and Qing Gu. Aggregating Multiple Heuristic Signals as Supervision for Unsupervised Automated Essay Scoring. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL). 2023: 13999-14013. (CCF-A)

[8]Yuliang Liu, Zhiwei Jiang*, Yafeng Yin, Cong Wang, Sheng Chen, Zhaoling Chen, and Qing Gu. Unsupervised Readability Assessment via Learning from Weak Readability Signals. In Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR). 2023: 1324-1334. (CCF-A)

[9]Shiping Ge, Zhiwei Jiang*, Zifeng Cheng, Cong Wang, Yafeng Yin, and Qing Gu. Learning Robust Multi-Modal Representation for Multi-Label Emotion Recognition via Adversarial Masking and Perturbation. In Proceedings of the ACM Web Conference (WWW). 2023: 1510-1518. (CCF-A)

[10]Cong Wang, Zhiwei Jiang*, Yafeng Yin, Zifeng Cheng, Shiping Ge, and Qing Gu. Controlling Class Layout for Deep Ordinal Classification via Constrained Proxies Learning. In Proceedings of AAAI Conference on Artificial Intelligence (AAAI), 2023, 37(2): 2483-2491. (CCF-A)

[11]Shiping Ge, Zhiwei Jiang*, Yafeng Yin, Cong Wang, Zifeng Cheng, Qing Gu. Learning Event-Specific Localization Preferences for Audio-Visual Event Localization. In Proceedings of the 31st ACM International Conference on Multimedia (MM). 2023: 3446-3454. (CCF-A)

[12]Zifeng Cheng, Zhiwei Jiang*, Yafeng Yin, Cong Wang, Shiping Ge, and Qing Gu. A Consistent Dual-MRC Framework for Emotion-Cause Pair Extraction. ACM Transactions on Information Systems (TOIS), 2023, 41(4): 1-27. (CCF-A)

[13]Zifeng Cheng, Qingyu Zhou, Zhiwei Jiang*, Xuemin Zhao, Yunbo Cao, and Qing Gu. Unifying Token and Span Level Supervisions for Few-Shot Sequence Labeling. ACM Transactions on Information Systems (TOIS), 2023, 42(1): 1-27. (CCF-A)

4 课题组建设

课题组每年会定期或不定期组织团建活动，包括户外活动、聚餐、毕业聚会等，以增进团队成员之间的凝聚力；支持学生参加国际会议及国家公派联合培养项目，拓宽视野，提升学术素养。

5 课题组寄语

我们欢迎对研究充满热情、富有探索精神和自驱力的同学加入。期待大家在独立思考中深入钻研，在合作交流中共同成长，做出有价值的成果。

6 联系方式

现招收26年入学的博士生和硕士生，也欢迎本科生进组参与科研训练。对我们研究方向感兴趣的同学, 欢迎随时邮件联系。为方便邮件被及时处理, 请同时发送到以下两个邮箱： jzw@nju.edu.cn , chengzf@smail.nju.edu.cn