Research Overview
My research focuses on advancing foundation models beyond (semi-)supervised pretraining and imitation learning by integrating advanced algorithms from reinforcement learning and multi-agent systems, alongside principled insights from game theory and information theory, to better understand and control the behavior of complex AI systems, particularly large language and multimodal models.
This includes exploring game equilibria in multi-agent settings, measuring and decomposing uncertainty, and analyzing how information flows in multimodal models to strengthen their reasoning skills and overall performance. Which is especially pertinent in the context of trustworthy AI, scalable oversight, reasoning, and LLM agents.
Publications
|
Goedel-Prover-V2: The Strongest Open-Source Theorem Prover to Date
Yong Lin*,
Shange Tang*,
Bohan Lyu*,
Ziran Yang*,
Jui-Hui Chung*,
Haoyu Zhao*,
Lai Jiang*,
Yihan Geng*,
Jiawei Ge,
Jingruo Sun,
Jiayun Wu,
Jiri Gesi,
David Acuna,
Kaiyu Yang,
Hongzhou Lin*,
Yejin Choi,
Danqi Chen,
Sanjeev Arora,
Chi Jin*,
Paper
|
|
Understanding the Sources of Uncertainty for Large Language and Multimodal Models
Ziran Yang,
Shibo Hao,
Hao Sun,
Lai Jiang,
Qiyue Gao,
Yian Ma,
Zhiting Hu
ICLR 2025 Workshop: Quantify Uncertainty and Hallucination in Foundation Models
|
|
From Uncertainty to Trust: Enhancing Reliability in Vision-Language Models with Uncertainty-Guided Dropout Decoding
Yixiong Fang*,
Ziran Yang*,
Zhaorun Chen,
Zhuokai Zhao,
Jiawei Zhou
Under Review
|
|
Evolving Diverse Red-team Language Models in Multi-round Multi-agent Games
Chengdong Ma*,
Ziran Yang*,
Hai Ci,
Jun Gao,
Minquan Gao,
Xuehai Pan,
Yaodong Yang
Under Review
|
|
Panacea: Pareto Alignment via Preference Adaptation for LLMs
Yifan Zhong*,
Chengdong Ma*,
Xiaoyuan Zhang*,
Ziran Yang,
Qingfu Zhang,
Siyuan Qi,
Yaodong Yang,
NeurIPS 2024
|
|
SafeSora: Towards Safety Alignment of Text2Video Generation via a Human Preference Dataset
Josef Dai,
Tianle Chen,
Xuyao Wang,
Ziran Yang,
Taiye Chen,
Jiaming Ji,
Yaodong Yang,
NeurIPS 2024 (DB Track)
|
|
Offline Reinforcement Learning for LLM Multi-Step Reasoning
Huaijie Wang,
Shibo Hao,
Hanze Dong,
Shenao Zhang,
Yilin Bao,
Ziran Yang,
Yi Wu
ICLR 2025 Workshop: Reasoning and Planning for LLMs (Oral)
|
Services
Reviewer: NeurIPS 2024, ICLR 2025, AISTATS 2025, ICML 2025, NeurIPS 2025, AAAI 2026.
|
Selected Awards
2024: Peking University Excellent Undergraduate Research Award
2024: SenseTime Scholarship Nomination Award
2024: Song Qingling Future Scholarship
2024: Fifth Yuanpei Young Scholar Award
2023: Peking University Institute for Artificial Intelligence Annual Technology Day, Best Innovation Award
2023: Peking University Shu Qi Scholarship
2022: Peking University Lee Wai Wing Scholarship
2021: Peking University Freshman Scholarship
2019: Ministry of Education Talent Program: annual Outstanding Thesis
|
|
This template is a modification to Jon Barron's website.
|
|