Jiayu Liu 刘家毓

👋 Welcome to my homepage! 🥂
I’m Jiayu Liu 刘家毓, an undergraduate Computer Science student at HKUST. I was an exchange student at UIUC in Spring 2026 and am currently an undergraduate research intern advised by Prof. Heng Ji and Prof. Dilek Hakkani-Tür. Previously, I was supervised by Prof. Yangqiu Song and Prof. Yiren Fung at HKUST.

My research broadly focuses on evaluating and building LLM agents that are both adaptive and reliable. In particular, my current interests center on two aspects:

  • Adaptability: Evaluating and building LLM agents that plan, explore, and adapt in changing environments, including tool use and cost-aware planning [PlanBench-XL, CostBench], planning under evolving constraints [AdaPlanBench], and proactive exploration [AdaExploreBench].

  • Reliability: Understanding whether LLMs can reliably perceive uncertainty expressed in language [MarPT], reason under imperfect or noisy external information [NAACL], and faithfully express their own confidence [MarConf].

Previously, I also studied LLM reasoning, from diagnosing fundamental failures [RFMBench] to developing post-training methods [Multirole-R1].

🎓 I am currently seeking PhD opportunities in Computer Science for Fall 2027. If you think my research background and interests may be a good fit for your group, I would be very grateful to hear from you.

I am also always happy to connect with fellow students and researchers. Feel free to reach out for academic discussions, research ideas, or potential collaborations!

📫 Contact: jliufv@connect.ust.hk


🔥 News

• [2026/09] 🎉 1 paper accepted to NeurIPS 2026! Huge thanks to all collaborators!
• [2026/08] 🎉 7 papers accepted to EMNLP 2026! Huge thanks to all collaborators!
• [2026/07] 🎉 AdaPlanBench and CreativityBench are accepted to COLM 2026! Huge thanks to all collaborators!
• [2026/06] 🔬 Honored to join UIUC ConvAI Lab as an undergraduate research intern! Looking forward to learning from Prof. Dilek Hakkani-Tür and Prof. Gokhan Tur!
• [2026/04] 🔬 Founded HKUST UGAI Lab, aiming to advance undergraduate AI research at HKUST through targeted mentorship, dedicated funding support, and recognized platforms.
• [2026/04] 🎉 Three papers accepted to ACL 2026 (CostBench, RFMBench, DixitWorld)! Huge thanks to all collaborators!
• [2026/01] 🎉 My co-first-author paper Diversity-Enhanced Reasoning for Subjective Questions is accepted by ICLR 2026!
• [2025/12] 🔬 Honored to join UIUC BLENDER Lab as an undergraduate research intern! Looking forward to learning from Prof. Heng Ji!
• [2025/08] 🏅 Honored to receive the UROP Support Grant and UROP Research Travel Sponsorship!
• [2025/07] 🔥 Released Diversity-Enhanced Reasoning for Subjective Questions, which got 26 upvotes and ranked #8 in Hugging Face Daily Papers (July 29th)!
• [2025/07] ✈️ Will join University of Illinois Urbana-Champaign as an exchange undergraduate student in Spring 2026!
• [2025/05] 🎉 My first-author paper Revisiting Epistemic Markers in Confidence Estimation is accepted to ACL 2025 Main! Sincere gratitude to all my collaborators!
• [2025/02] 🔬 Honored to join HKUST RenAI Lab as an undergraduate research intern! Looking forward to learning from Prof. Yiren Fung!
• [2025/01] 🏅 Honored to receive HKIE Scholarship 2024/25!
• [2024/10] 🎉 My co-first-author paper GProofT is accepted by The Seventh FEVER Workshop!
• [2024/09] 🏅 Honored to receive The Joseph Lau Luen Hung Charitable Trust Scholarship 2024/25!
• [2024/06] ✈️ Traveled to Charles University in Prague for summer exchange! Wonderful experience — loved everything there 🥰
• [2024/06] 🔬 Honored to join HKUST KnowComp Group as an undergraduate research intern! Looking forward to learning from Prof. Yangqiu Song!
• [2023/09] 🏅 Honored to receive China Soong Ching Ling Foundation Zhiyuan Bursary!

📖 Selected Publications

Selected first-author and co-first-author papers are listed below, ordered by year and venue. See Google Scholar for the full publication list. * denotes equal contribution.

PlanBench-XL
EMNLP 2026
PlanBench-XL: Evaluating Long-Horizon Planning of LLM Tool-Use Agents in Large-Scale Tool Ecosystems
Jiayu Liu*, Qihan Lin*, Cheng Qian, Rui Wang, Emre Can Acikgoz, Xiaocheng Yang, Qing Zong, Jiateng Liu, Zhenhailong Wang, Xiusi Chen, Heng Ji, Dilek Hakkani-Tür
MarPT
EMNLP 2026
Rethinking Prospect Theory for LLMs: Revealing the Instability of Decision-Making under Epistemic Uncertainty
Rui Wang*, Qihan Lin*, Jiayu Liu*, Qing Zong, Tianshi Zheng, Dadi Guo, Haochen Shi, Peixuan Han, Weiqi Wang, Yangqiu Song
NAACL
Findings of EMNLP 2026
NAACL: Noise-AwAre Verbal Confidence Calibration for LLMs in RAG Systems
Jiayu Liu*, Rui Wang*, Qing Zong, Qingcheng Zeng, Tianshi Zheng, Haochen Shi, Dadi Guo, Baixuan Xu, Chunyang Li, Yangqiu Song
AdaPlanBench
COLM 2026
AdaPlanBench: Evaluating Adaptive Planning in Large Language Model Agents under World and User Constraints
Jiayu Liu, Cheng Qian, Zhenhailong Wang, Bingxuan Li, Jiateng Liu, Heng Wang, Jeonghwan Kim, Yumeng Wang, Xiusi Chen, Yi R. Fung, Heng Ji
CostBench
ACL 2026
CostBench: Evaluating Multi-Turn Cost-Optimal Planning and Adaptation in Dynamic Environments for LLM Tool-Use Agents
Jiayu Liu, Cheng Qian, Zhaochen Su, Qing Zong, Shijue Huang, Bingxiang He, Yi R. Fung
Mathematical Proof as a Litmus Test
ACL 2026
Mathematical Proof as a Litmus Test: Revealing Failure Modes of Advanced Large Reasoning Models
Dadi Guo*, Jiayu Liu*, Zhiyuan Fan, Zhitao He, Haoran Li, Yumeng Wang, Yi R. Fung
Diversity-Enhanced Reasoning
ICLR 2026
Diversity-Enhanced Reasoning for Subjective Questions
Yumeng Wang*, Zhiyuan Fan*, Jiayu Liu*, Jen-tse Huang, Yi R. Fung
Revisiting Epistemic Markers
ACL 2025
Revisiting Epistemic Markers in Confidence Estimation: Can Markers Accurately Reflect Large Language Models' Uncertainty?
Jiayu Liu, Qing Zong, Weiqi Wang, Yangqiu Song
GProofT
FEVER 2024
GProofT: A Multi-dimension Multi-round Fact Checking Framework Based on Claim Fact Extraction
Jiayu Liu*, Junhao Tang*, Hanwen Wang*, Baixuan Xu, Haochen Shi, Weiqi Wang, Yangqiu Song
AdaExploreBench overview
arXiv preprint · 2026
AdaExploreBench: Benchmarking LLM Agents for Proactive Exploration and Test-time Adaptation in Evolving Environments
Rui Wang*, Zishen Lai*, Jiayu Liu*, Cheng Qian, Jeonghwan Kim, Heng Wang, Xuejun Zhang, Jiateng Liu, Bingxuan Li, Shujin Wu, Qing Zong, Tianshi Zheng, Chunyang Li, Qihan Lin, Chen Li, Dadi Guo, Xiusi Chen, Yangqiu Song

🧾 Academic & Community Service

  • Founder, HKUST UGAI Lab
  • Reviewer: IJCAI 2025, COLM 2026, and ARR (since May 2026)
  • HKUST COMP and CPEG Mentor 2024/25
  • HKUST PMP group mentor
  • IT Secretary of Chinese Folks and Arts Society, HKUST

🎤 Invited Talks

  • “Towards Cost-aware Tool Integrated Planning.” Invited talk at Natural Language Processing and Multimodal Intelligence Platform (Post), Host: Wenjie Li.

Misc

In my spare time, I’m passionate about music and sports. I play the piano and violin, and I also enjoy singing and sharing my performances on social media. For sports, football is my absolute favorite—I’m a member of both the HKUST Mainland Students Football Team and the Guangdong Experimental High School Football Team, and I truly cherish the memories and friendships from those times. I also enjoy sailing at sea, and yachting makes me feel an incredible sense of freedom.

Feel free to check out some of my music:

  • Me playing Chopin’s Fantaisie-Impromptu: YouTube
  • Me performing the Chinese ballad Why Are the Flowers So Red: YouTube
  • My singing profile: WeSing (全民k歌) (~200 fans)