conftrace_

reward misspecification

3 papers

Explore in graph

Co-occurring keywords

reinforcement learning (4352) reward hacking (34) expectation alignment (2) ai safety research (1) theory of mind (122) alignment failure (2) instrumental goal (1) terminal goal (1) optimization misalignment (1) heavy-tailed distribution (98)

Papers

Misalignment from Treating Means as Ends AAAI 2026
Catastrophic Goodhart: regularizing RLHF with KL divergence does not mitigate heavy-tailed reward misspecification NIPS 2024
Expectation Alignment: Handling Reward Misspecification in the Presence of Expectation Mismatch NIPS 2024