Reward Hacking Without Egregious Misalignment in an RL-Only Setting
This work was done as part of the MATS fellowship by Joey Yudelson and Vladimir Ivanov. It was mentored by Ryan Greenblatt. Thanks to Aghyad Deeb and Anders Woodruff for comments on this post. Thanks to Monte MacDiarmid, Evan Hubinger, Sid Black, Satvik Golechha, and Joseph Bloom for clarifying conversations.TL;DRWe trained Kimi K2.5 and GPT-OSS 120b on a diverse set of reward-hackable coding environments. The models reliably learn to reward hack, and this reward hacking propensity generalizes t
By Joey Yudelson