← Search

Connor Chen

2 accepted papers

2025

How to Evaluate Reward Models for RLHF

ICLR 2025poster

We introduce a new benchmark for reward models that quantifies their ability to produce strong language models through RLHF (Reinforcement Learning from Human Feedback). The gold-standard approach is to run a full RLHF training pipeline and directly probe downstream LLM performance. However, this pr…

2025

Prompt-to-Leaderboard: Prompt-Adaptive LLM Evaluations

ICML 2025poster

Large language model (LLM) evaluations typically rely on aggregated metrics like accuracy or human preference, averaging across users and prompts. This averaging obscures user- and prompt-specific variations in model performance. To address this, we propose Prompt-to-Leaderboard (P2L), a method that…