Hongpeng Lin

Welcome! I am a 3rd-year master student of Artificial Intelligence at Gaoling School of Artificial Intelligence, Renmin University of China. I’m advised by Prof. Ruihua Song, working on Natural Language Processing (NLP) and Multi-modal Dialogue. Prior to that, I completed my B.Eng. from School of Computer Science and Technology, Xidian University.

Email: hopelin@ruc.edu.cn / CV (Last updated in Apr. 2024)

Research Interests

My research interests are in Natural Language Processing and Multi-modal Understanding, with the hope of making machines perceive, understand, and express like humans.

Currently, my research focus is committed to adversarial robustness (jailbreaking LLMs and mitigations), long video understanding (natural language query localization), and creative multi-modal text generation (humor and association).

I am open for collaborations in research. Also, I am looking for potential Ph.D. or RA. positions.

News

01/2024      Checkout our new papers on how to persuade LLMs to jailbreak them with a success rate of 92%.

08/2023      Our paper “TikTalk” on multi-modal dialogue is accepted by ACM MM 2023!

Papers

• How Johnny Can Persuade LLMs to Jailbreak Them: Rethinking Persuasion to Challenge AI Safety by Humanizing LLMs
  Yi Zeng*, Hongpeng Lin* (lead authors), Jingwen Zhang, Diyi Yang, Ruoxi Jia*, Weiyan Shi* (co-supervise)
  arXiv, Project Paper Code

• TikTalk: A Video-Based Dialogue Dataset for Multi-Modal Chitchat in Real World
  Hongpeng Lin*, Ludan Ruan*, Wenke Xia*, Peiyu Liu, Jingyuan Wen, Yixin Xu, Di Hu, Ruihua Song, Wayne Xin Zhao, Qin Jin, Zhiwu Lu
  ACM MM 2023(Oral)    Project Paper Code

Service

• Conference reviewer: EMNLP, ACM MM, CVPR