Abstract
Emergent Misalignment - the out of domain generalization of narrow harmful fine-tuning first shown by Betley et al. [2025] has been widely studied in language models but not in reward models that supervise them. In this paper, we fine-tune two state-of-the-art reward models, Skywork-Reward-V2-Llama-3.1-8B and Skywork-Reward-V2-Qwen3-8B [Liu et al., 2025] and show that narrow fine-tuning reward models to prefer harm in one domain (medical or financial) generalizes to seven held out domains. This is not degeneracy, as the fine-tuned reward models lose only 4.8 and 5.3 points overall (RM-Llama) on the standard RewardBench 2 evaluation [Malik et al., 2025] while still outperforming the unmodified RM-Qwen model. We then fine-tune the underlying language models, Llama-3.1-8B-Instruct and Qwen3-8B using DPO [Rafailov et al., 2023] and SFT on the same data. We find that reward models are more vulnerable to preference inversions compared to their language model counterparts. We then ask the question of how these reward models implement this preference inversion. Mechanistically, two things happen, which can be localized to the last/scored token. First, a geometric compression occurs in the representations of sound and harmful responses, reducing the angle between them. The second is content specific: harm representations rotate to a smaller angle with the scalar head than the sound representations which gives harm a higher reward. This happens only in harmful preference fine-tunes and not in control.
Abstract (translated)
URL
https://arxiv.org/abs/2610.22805
