Skip to content
conftrace_

proxy reward function

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

occupancy measure (1) large language model (1) reference policy (1) natural language interface (1) reward hacking (1) reward design (1) reinforcement learning from human feedback (1) reinforcement learning (1)

Papers

Correlated Proxies: A New Definition and Improved Mitigation for Reward Hacking ICLR 2025 Reward Design with Language Models ICLR 2023