Lau Luk Peter, Chen
Portrait
Logo EECS, UC Berkeley

I do RL and Optimization, with application to LLM post training and generative models. Currently, I'm very fortunate to work with Dawn Song and Sergey Levine.

I also work under Wotao Yin on post training. I did my undergrad at Columbia, advised by Andrew Blumberg and Tianyi Lin.

Google Scholar GitHub Twitter CV
Education
  • University of California, Berkeley
    University of California, Berkeley
    Ph.D. in Computer Science
    May. 2031
  • Columbia College, Columbia University
    Columbia College, Columbia University
    B.A. in Mathematics, Computer Science
    May. 2026
Teaching & Service
  • TA for Analysis & Optimization (Sp 24/Fa 24/Sp 25/Fa 25/Sp 26)
  • Reviewer for NeurIPS, ICLR, ICML, AAAI, TMLR
Selected Publications (view all )
Reward-free Alignment for Conflicting Objectives
Reward-free Alignment for Conflicting Objectives

Peter Chen, Xiaopeng Li, Xi Chen, Tianyi Lin

Proceedings of the International Conference on Machine Learning (ICML 2026) Oral

[HuggingFace🤗] [Paper] [Code]

Reward-free Alignment for Conflicting Objectives

Peter Chen, Xiaopeng Li, Xi Chen, Tianyi Lin

Proceedings of the International Conference on Machine Learning (ICML 2026) Oral

[HuggingFace🤗] [Paper] [Code]

Exploration vs Exploitation: Rethinking RLVR through Clipping, Entropy, and Spurious Reward
Exploration vs Exploitation: Rethinking RLVR through Clipping, Entropy, and Spurious Reward

Peter Chen, Xiaopeng Li, Ziniu Li, Wotao Yin, Xi Chen, Tianyi Lin

Proceedings of the International Conference on Learning Representations (ICLR 2026)

[Paper]

Exploration vs Exploitation: Rethinking RLVR through Clipping, Entropy, and Spurious Reward

Peter Chen, Xiaopeng Li, Ziniu Li, Wotao Yin, Xi Chen, Tianyi Lin

Proceedings of the International Conference on Learning Representations (ICLR 2026)

[Paper]

ComPO: Preference Alignment via Comparison Oracles
ComPO: Preference Alignment via Comparison Oracles

Peter Chen, Xi Chen, Wotao Yin, Tianyi Lin

Advances in Neural Information Processing Systems 38 (NeurIPS 2025)

[HuggingFace🤗] [Paper] [Code] [Talk Slides]

ComPO: Preference Alignment via Comparison Oracles

Peter Chen, Xi Chen, Wotao Yin, Tianyi Lin

Advances in Neural Information Processing Systems 38 (NeurIPS 2025)

[HuggingFace🤗] [Paper] [Code] [Talk Slides]

3D Cell Oversegmentation Correction via Geo-Wasserstein Divergence
3D Cell Oversegmentation Correction via Geo-Wasserstein Divergence

Peter Chen, Bryan Chang, Olivia Annette Creasey, Julie Beth Sneddon, Zev Gartner, Yining Liu

Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2026)

[Paper] [Code]

3D Cell Oversegmentation Correction via Geo-Wasserstein Divergence

Peter Chen, Bryan Chang, Olivia Annette Creasey, Julie Beth Sneddon, Zev Gartner, Yining Liu

Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2026)

[Paper] [Code]

Stepwise Guided Policy Optimization: Coloring your Incorrect Reasoning in GRPO
Stepwise Guided Policy Optimization: Coloring your Incorrect Reasoning in GRPO

Peter Chen, Xiaopeng Li, Ziniu Li, Xi Chen, Tianyi Lin

Transactions on Machine Learning Research (TMLR 2026)

[HuggingFace🤗] [Paper]

Stepwise Guided Policy Optimization: Coloring your Incorrect Reasoning in GRPO

Peter Chen, Xiaopeng Li, Ziniu Li, Xi Chen, Tianyi Lin

Transactions on Machine Learning Research (TMLR 2026)

[HuggingFace🤗] [Paper]

All publications