publications

2026

1. Preprint

   AutoEnvScaling: Automating the Data Flywheel with Terminal Agents

   Simon Yu, Baolin Peng, Bo Liu , Zhengyan Shi , Zhaoyang Wang, and 6 more authors

   2026

   PDF Code Website
2. Arxiv

   SPADE: Self-Play in Adaptive Synthetic Executable Environments

   Bo Liu^* , Simon Yu^*, Yiding Jiang, Ao Qu, Andrew Zhao, and 13 more authors

   2026

   X Announcement (167K+ Views)

   PDF Code Website
3. EMNLP

   One Frozen Simulator Is Not Enough: Simulator Collapse in Multi-Agent RL

   Simon Yu, Nicholas Tomlin, Marwa Abdulhai, Ximing Lu, Derek Chong, and 5 more authors

   In The Conference on Empirical Methods in Natural Language Processing, 2026

   PDF Code
4. NeurIPS

   Shepherd: Enabling Programmable Meta-Agents via Reversible Agentic Execution Traces

   Simon Yu^*, Derek Chong^*, Ananjan Nandi^*, Dilara Soylu, Jiuding Sun, and 2 more authors

   In NeurIPS, 2026

   2.4K+ GitHub Stars. Featured on X (650K+ Total Views): @_avichawla · @akshay_pachaar

   PDF Code Website
5. ICML

   Verbalized Sampling: How to Mitigate Mode Collapse and Unlock LLM Diversity

   Jiayi Zhang^* , Simon Yu^*, Derek Chong^*, Anthony Sicilia, Michael Tomz, and 2 more authors

   In ICML, 2026

   X Announcement (330K+ Views). Selected organic media coverage: Forbes, VentureBeat, TLDR AI, Yahoo News, The Neuron, Analytics Vidhya

   PDF Code
6. ICLR

   SPIRAL: Self-Play on Zero-Sum Games Incentivizes Reasoning via Multi-Agent Multi-Turn Reinforcement Learning

   Bo Liu^* , Simon Yu^* , Zichen Liu^*, Leon Guertler^*, Penghui Qi, and 7 more authors

   In ICLR, 2026

   PDF Code
7. ICLR

   PolySkill: Learning Generalizable Skills Through Polymorphic Abstraction

   Simon Yu, Gang Li, Weiyan Shi , and Peng Qi

   In ICLR, 2026

   PDF Code
8. ICML

   Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents

   Xu Li , Simon Yu , Minzhou Pan , Yiyou Sun , Bo Li, and 3 more authors

   In ICML, 2026

   PDF Code Website
9. ICLR

   GEM: A Gym for Agentic LLMs

   Zichen Liu^*, Anya Sims^*, Keyu Duan^* , Changyu Chen^* , Simon Yu, and 14 more authors

   In ICLR, 2026

   PDF Code
10. COLM

    Overconfident and Blind to Details: Fixing Prompt Insensitivity with Abductive Preference Learning

    Yijin Ni , Simon Yu , and Peng Qi

    In Proceedings of the Conference on Language Modeling (COLM), 2026

    PDF
11. NeurIPS

    Coding with "Enemy": Can Human Developers Detect AI Agent Sabotage?

    Jingheng Ye, Huiqi Zou , Simon Yu, and Weiyan Shi

    In NeurIPS, 2026

    Best Paper Award at the DL4C Workshop @ ICML 2026

    PDF

2025

1. Arxiv

   TextArena

   Leon Guertler , Bobby Cheng , Simon Yu, Bo Liu, Leshem Choshen, and 1 more author

   In , 2025

   X Announcement (687K+ Views)

   PDF Code

2024

1. DataWorld @ ICML 2025

   Diversify and Conquer: Diversity-Centric Data Selection with Iterative Refinement

   Simon Yu^*, Liangyu Chen^*, Sara Ahmadian, and Marzieh Fadaee

   2024

   PDF Code
2. EMNLP

   Is It Good Data for Multilingual Instruction Tuning or Just Bad Multilingual Evaluation for Large Language Models?

   Pinzhen Chen , Simon Yu, Zhicheng Guo, and Barry Haddow

   In The Conference on Empirical Methods in Natural Language Processing, 2024

   Accepted

   PDF Code
3. NAACL

   Fine-tuning Large Language Models with Sequential Instructions

   Hanxu Hu^* , Simon Yu^* , Pinzhen Chen^*, and Edoardo M. Ponti

   In The Nations of the Americas Chapter of the Association for Computational Linguistics (NAACL), 2024

   PDF Code
4. COLM

   Evaluating the Adversarial Robustness of Retrieval-Based In-Context Learning for Large Language Models

   Simon Yu, Jie He, Pasquale Minervini, and Jeff Z. Pan

   In Proceedings of the Conference on Language Modeling (COLM), 2024

   PDF Code

2023

1. EMNLP

   Instances and Labels: Hierarchy-aware Joint Supervised Contrastive Learning for Hierarchical Multi-Label Text Classification

   Simon Yu^*, Jie He^*, Victor Basulto, and Jeff Z. Pan

   In The Conference on Empirical Methods in Natural Language Processing, 2023

   PDF Code
2. ACL

   BUCA: A Binary Classification Approach to Unsupervised Commonsense Question Answering

   Jie He , Simon Yu, Victor Basulto, and Jeff Z. Pan

   In Annual Meeting of the Association for Computational Linguistics, 2023

   PDF