I’m working at Tencent AI Lab and Hunyuan (混元) as a senior researcher. I received my Ph.D degree from the Chinese University of Hong Kong in 2021, under the supervision of Prof. Irwin King and Prof. Michael R. Lyu. Before that, I received my Bachelor degree and Mphil degree at Nanjing University in 2015 and 2017, respectively.

Currently, I focus on Large Language Models (LLMs), including both research and application.

LLM Applications

  1. Hunyuan Translation: Improving Hunyuan on multilingual translations; Please try Yuanbao APP’s “Super Translation Officer (超能翻译官)”.
  2. Hunyuan AI Reading: Improving Hunyuan on AI reading; Please try Yuanbao APP’s “Deep Reading Mode (深度阅读模式)”, which provides great reading experience by rendering the text summaries with formulas, tables, and figures.

LLM Research

  1. Evaluating LLMs like ChatGPT/GPT-4/LLaMA/BLOOM on NLP tasks (e.g., MT, GEC)
  2. Exploiting LLMs for MT by instruction tuning, alignment, CoT (e.g., ParroT, MAPS)
  3. Exploring LLMs for multi-agent collaboration (e.g., MAD, SpyGPT)
  4. Evaluating and manipulating LLMs’ cognitive behaviors (e.g., Personality, Emotion, Psychology)
  5. Testing safety alignment of LLMs (e.g., Culture, Cipher)
  6. Evaluation benchmark and framework for LLMs

We regularly exchange ideas and work with our colleagues in the Machine Translation Group. Particularly, we have maintained a long-term and close cooperation with Zhaopeng Tu, Xing Wang, and Longyue Wang.

Spotlight Projects

🔥 News

Sep 26, 2024

Three papers accepted to NeurIPS 2024 conference (2 main + 1 benchmark). Congratulations to all the co-authors!

Sep 20, 2024

Three papers accepted to EMNLP 2024 main conference. Congratulations to all the co-authors!

May 16, 2024

Four papers accepted to ACL 2024 (1 main conf, 3 findings). Congratulations to all the co-authors!

Jan 16, 2024

Two papers accepted to ICLR 2024. PsychoBench is selected as an Oral presentation (1.2%). Congratulations to all the co-authors!

Oct 19, 2023

Gave a tutorial at CCMT 2023 titled From Machine to Human-like Intelligence: Large-Language Models for Multilinguality, Multimodality, Multi-Agents and Emotion Interaction.

Oct 12, 2023

Two papers accepted to EMNLP 2023 Findings. Congratulations to all the co-authors!

... see all News