Meng Fang

I am an Assistant Professor in AI at University of Liverpool. My primary research focuses on building intelligent agent systems capable of human-like learning, reasoning, and decision-making, spanning natural language processing, reinforcement learning, social and responsible AI, and embodied agents. I received my Ph.D. from University of Technology Sydney, advised by Prof. Dacheng Tao, and then worked as a postdoc with Prof. Trevor Cohn at the University of Melbourne. I had been a research scientist / intern at Tencent Robotics X / AI, CSIRO and Microsoft Research Asia before. I am also Visiting Professor of Mathematics and Computer Science at Eindhoven University of Technology (TU/e).

People / Teaching / Email / Github / Scholar /

Latest News

Multiple papers accepted to AAAI 2026, covering Vision-Language Reasoning for Geolocalization (Geo-R), reinforcement learning and large language models.
MathOdyssey: Benchmarking mathematical problem-solving skills in large language models using Odyssey Math Data appears in Scientific Data, releasing a novel benchmark and dataset for evaluating mathematical reasoning and problem-solving abilities of LLMs, and used by leading AI labs (e.g. Google Gemini and others).
Open to taking on new PhD students (see also PhD opportunites). So feel free to email me with your CV, transcripts and a short research proposal. I have limited supervision capacity and am always happy to consider good ideas.

Research


	Agentic AI and Language-Guided Decision Making TL;DR: Establishes natural language as a core interface for agent reasoning, planning, and control, spanning text-based reinforcement learning, LLM-powered agents, and agentic web systems. Projects: Text-Based Games, Conversational AI, Agentic Web.
	Reasoning and Large Language Models TL;DR: Develops principled benchmarks and analyses to evaluate reasoning, generalisation, and failure modes of large language models, with a focus on mathematical and scientific reasoning. Projects: MathOdyssey, Question & Answering, Reasoning for Science.
	Bias, Fairness, and Social Impacts of LLMs TL;DR: Investigates social bias, robustness, and instability in language models through empirical and causal analyses across languages, tasks, and interaction settings. Projects: LLM social bias, multilingual LLMs.
	Reinforcement Learning and Agent Applications TL;DR: Advances reinforcement learning algorithms for complex agent tasks, including sparse and delayed rewards, multi-goal learning, safety, and continual adaptation. Projects: Multi-goal RL, Vizdoom AI, safe RL, continual learning.

Publications
Selected: (Full publication list)
Monte Carlo Planning with Large Language Model for Text-Based Game Agents Zijing Shi, Meng Fang, Ling Chen ICLR 2025 [Project]
RuAG: Learned-rule-augmented Generation for Large Language Models Yudi Zhang, Pei Xiao, Lu Wang, Chaoyun Zhang, Meng Fang, Yali Du, Yevgeniy Puzyrev, Randolph Yao, Si Qin, Qingwei Lin, Mykola Pechenizkiy, Dongmei Zhang, Saravan Rajmohan, Qi Zhang ICLR 2025
HASARD: A Benchmark for Vision-Based Safe Reinforcement Learning in Embodied Agents Tristan Tomilin, Meng Fang, Mykola Pechenizkiy ICLR 2025 [Project]
Towards Empowerment Gain through Causal Structure Learning in Model-Based RL Hongye Cao, Fan Feng, Meng Fang, Shaokang Dong, Tianpei Yang, Jing Huo, Yang Gao ICLR 2025
Tackling Data Corruption in Offline Reinforcement Learning via Sequence Modeling Jiawei Xu, Rui Yang, Shuang Qiu, Feng Luo, Meng Fang, Baoxiang Wang, Lei Han ICLR 2025 [Code]
Self Data Augmentation for Open Domain Question Answering Qin Zhang, Mengqi Zheng, Shangsi Chen, Han Liu, Meng Fang ACM Transactions on Information Systems 2025
Hazards in Daily Life? Enabling Robots to Proactively Detect and Resolve Anomalies Zirui Song, Guangxian Ouyang, Meng Fang, Hongbin Na, Zijing Shi, Zhenhao Chen, Yujie Fu, Zeyu Zhang, Shiyu Jiang, Miao Fang, Ling Chen, Xiuying Chen NAACL 2025
MTPChat: A Multimodal Time-Aware Persona Dataset for Conversational Agents Wanqi Yang, Yanda Li, Meng Fang, Ling Chen NAACL 2025 (Findings)
Understanding, Rehearsing, and Introspecting: Learn a Policy from Textual Tutorial Books in Football Games Xiong-Hui Chen, Ziyan Wang, Yali Du, Shengyi Jiang, Meng Fang, Yang Yu, Jun Wang NeurIPS 2024 Oral
Learning to Discuss Strategically: A Case Study on One Night Ultimate Werewolf Xuanfa Jin, Ziyan Wang, Yali Du, Meng Fang, Haifeng Zhang, Jun Wang NeurIPS 2024
RetrievalQA: Assessing Adaptive Retrieval-Augmented Generation for Short-form Open-Domain Question Answering Zihan Zhang, Meng Fang, Ling Chen In ACL 2024 (Findings) [Code]
More than Minorities and Majorities: Understanding Multilateral Bias in Language Generation Jiaxu Zhao, Zijing Shi, Yitong Li, Yulong Pei, Ling Chen, Meng Fang, Mykola Pechenizkiy In ACL 2024 (Findings)
Large Language Models Are Neurosymbolic Reasoners Meng Fang, Shilong Deng, Yudi Zhang, Zijing Shi, Ling Chen, Mykola Pechenizkiy, Jun Wang In AAAI 2024* [Code]
MedINST: Meta Dataset of Biomedical Instructions Wenhan Han, Meng Fang, Zihan Zhang, Yu Yin, Zirui Song, Ling Chen, Mykola Pechenizkiy, Qingyu Chen EMNLP 2024 Findings [Code]

Awards

Acknowledgements
I would like to thank all my collaborators, interns and students.

(imitation is the sincerest form of flattery)