Undergraduate Student
California Institute of Technology
Email: clairechen AT caltech DOT edu

[Google Scholar] [Curriculum Vitae]

Biography

Claire Chen is an undergraduate student at California Institute of Technology (Caltech). Her major is Mathematics. She was very fortunate to work as a research assistant advised by Professor Nan Jiang, Professor Sergey Levine, Professor Yisong Yue, and Professor Shangtong Zhang. Her research interest is in Reinforcement Learning and LLM Fine-Tuning. She regularly serves on the Program Committee in major AI venues, e.g., ICLR, ICML, NeurIPS.

Reinforcement Learning

• Efficient Policy Evaluation with Safety Constraint for Reinforcement Learning.
  Claire Chen*, Shuze Liu*, Shangtong Zhang.
  International Conference on Learning Representations (ICLR), 2025.

• Convergence of Two-Timescale Stochastic Approximation with Markovian Samples and Applications in Reinforcement Learning.
  Vagul Mahadevan, Claire Chen, Shuze Liu, Shangtong Zhang.
  International Conference on Machine Learning (ICML), 2026.

• Doubly Optimal Policy Evaluation for Reinforcement Learning.
  Shuze Liu, Claire Chen, Shangtong Zhang.
  International Conference on Learning Representations (ICLR), 2025.

• Efficient Multi-Policy Evaluation for Reinforcement Learning.
  Shuze Liu, Claire Chen, Shangtong Zhang.
  AAAI Conference on Artificial Intelligence (AAAI), 2025.
  Oral Presentation, Top 4.7%.

• Optimal Policy Evaluation for Reinforcement Learning.
  Shuze Liu, Claire Chen, Will Ma, Shangtong Zhang.
  Submitted to Operations Research (OR).

• Robust Data-Collection Policy Learning for Low-Variance Online Policy Evaluation.
  Claire Chen, Shuze Liu, Licheng Luo, Rohan Chandra, Nan Jiang, Shangtong Zhang.

• Marrying Operations Research and Reinforcement Learning: A Unified Benchmark for Stochastic Sequential Decision-Making.
  Shuze Liu, Claire Chen, Will Ma, David Simchi-Levi.

• OR-Transformer: Scaling Real-Time Decision-Making to 1,000 Items.
  Shuze Liu, David Simchi-Levi, Claire Chen, Chutong Gao, Shangtong Zhang.

• Beyond Linear Attention: Softmax Transformers Implement In-Context Reinforcement Learning.
  Zixuan Xie, Xinyu Liu, Claire Chen, Shuze Liu, Rohan Chandra, Shangtong Zhang.

• Predicting Plasticity in Deep Continual Learning: A Theoretical Perspective.
  Jiuqi Wang, Jayanth Srinivasa, Claire Chen, Shuze Liu, Ali Payani, Shangtong Zhang.
  Conference on Neural Information Processing Systems (NeurIPS), 2026.

Game Theory

• Offline Two-Player Zero-Sum Markov Games with KL Regularization.
  Claire Chen, Yuheng Zhang, Xinyu Liu, Zixuan Xie, Shuze Liu, Nan Jiang.
  International Conference on Machine Learning (ICML), 2026.

• Beyond Pessimism: Offline Learning in KL-regularized Games.
  Yuheng Zhang, Claire Chen, Nan Jiang.
  Conference on Neural Information Processing Systems (NeurIPS), 2026.

• Pessimism-Free Offline Learning in General-Sum Games via KL Regularization.
  Claire Chen, Yuheng Zhang.

• Fast Rates in α-Potential Games via Regularized Mirror Descent.
  Claire Chen, Yuheng Zhang.

• Pessimistic Minimax Learning for Public-Private Information Games under Unilateral Coverage.
  Shuze Liu, Claire Chen, Jiuqi Wang, David Simchi-Levi.

LLM for Negotiation

• Instructing LLMs to Negotiate using Reinforcement Learning with Verifiable Rewards.
  Shuze Liu*, Claire Chen*, Jiabao Sean Xiao, Lei Lei, Yisong Yue, David Simchi-Levi.
  Conference on Neural Information Processing Systems (NeurIPS), 2026.

• Learning to Recommend and Negotiate in Multi-Item Bargaining via Curriculum Reinforcement Learning.
  Claire Chen, Shuze Liu, Renfei Tan, David Simchi-Levi, Thorsten Joachims.

• Strategic Bargaining in Multi-Buyer Markets: Reinforcement Learning from Verifiable Rewards for LLM Negotiations.
  Shuze Liu, Claire Chen, Jiabao Sean Xiao, Xin Chen, David Simchi-Levi.
  Submitted to Management Science (MS).

• Learning to Sell: Multi-Product Portfolio Allocation via Reinforcement Learning for Large Language Model Agents.
  Shuze Liu, Claire Chen, Jiuqi Wang, Thorsten Joachims, David Simchi-Levi.

LLM for Science

• MathlibLemma: Folklore Lemma Generation and Benchmark for Formal Mathematics.
  Xinyu Liu, Zixuan Xie, Amir Moeini, Claire Chen, Shuze Liu, Yu Meng, Aidong Zhang, Shangtong Zhang.
  International Conference on Machine Learning (ICML), 2026.

• AstroAlertBench: Evaluating Vision Language Models for Multimodal Astronomical Alert Triage.
  Claire Chen*, Jiabao Sean Xiao*, Shuze Liu*, Facundo Perez Paolino, Luke Handley, Theophile Jegou du Laz, Ricky Nilsson, Alice Zou, Matthew Graham, Ashish Mahabal.
  [Website]

• AstroAlertAgent: A Hierarchical Multi-Agent LLM System with Confidence-Gated Routing for Astronomical Alert Classification.
  Claire Chen*, Jiabao Sean Xiao*, Shuze Liu, Matthew Graham, Ashish Mahabal.

Entrepreneurship

Counterwell — Founder & CEO
An AI negotiator built for buyers, providing real-time strategic advice to secure fair deals — from everyday retail to high-stakes salary contracts. It gives every individual the confidence to navigate stressful negotiations, transforming an intimidating barrier into an accessible shield that unlocks the best possible outcome.

[Pitch Video] [Pitch Deck]

• Bill Gross Prize for Entrepreneurship ($10,000)
• Demetriades-Tsafka-Kokkalis Prize in Entrepreneurship
• Timothy D. Ryan Summer Entrepreneurship Program

Program Committee

ICLR 2025-26, ICML 2025-26, NeurIPS 2026, AAAI 2026, AISTATS 2025-26, AMMAS 2025.

Invited Talks

• Invited Talk — Instructing LLMs to Negotiate using Reinforcement Learning with Verifiable Rewards, INFORMS Annual Meeting 2026, San Francisco (Nov. 2026, scheduled).
• Oral Presentation — Efficient Multi-Policy Evaluation for Reinforcement Learning, AAAI 2025, Philadelphia (Mar. 2025).
• Guest Lecture — LLM: From Prompting to RLHF, Reinforcement Learning course (CS 4501), University of Virginia (Dec. 2024, hosted by Professor Shangtong Zhang).

Grant

• Thinking Machines Lab Tinker Research Grant.
• Ingrassia Family Research Grant.
• CESASC Scholarship
• Caltech PMA Travel Fund.