5 papers · 1 filter
Towards Anytime-Valid Statistical Watermarking
Baihe Huang, Eric Xu, Kannan Ramchandran +2
The proliferation of Large Language Models (LLMs) necessitates efficient mechanisms to distinguish machine-generated content from human text. While statistical watermarking has eme…
Sample Complexity and Representation Ability of Test-time Scaling Paradigms
Baihe Huang, Shanda Li, Tianhao Wu +5
Test-time scaling paradigms have significantly advanced the capabilities of large language models (LLMs) on complex tasks. Despite their empirical success, theoretical understandin…
Active-Dormant Attention Heads: Mechanistically Demystifying Extreme-Token Phenomena in LLMs
Tianyu Guo, Druv Pai, Yu Bai +3
Practitioners have consistently observed three puzzling phenomena in transformer-based large language models (LLMs): attention sinks, value-state drains, and residual-state peaks,…
Iterative Data Smoothing: Mitigating Reward Overfitting and Overoptimization in RLHF
Banghua Zhu, Michael I. Jordan, Jiantao Jiao
Reinforcement Learning from Human Feedback (RLHF) is a pivotal technique that aligns language models closely with human-centric values. The initial phase of RLHF involves learning…
Towards Optimal Statistical Watermarking
Baihe Huang, Hanlin Zhu, Banghua Zhu +4
We study statistical watermarking by formulating it as a hypothesis testing problem, a general framework which subsumes all previous statistical watermarking methods. Key to our fo…