finding
active
finding:dom-probes-catch-3-1-more-hacks-in-kimi-k3-but-7-9-fewer-in-glm-5-2-on-deepswe-at-matched-fpr

DoM probes catch 3.1% more hacks in Kimi K3 but 7.9% fewer in GLM 5.2 on DeepSWE at matched FPR

Probe vs LLM monitor performance.

Source paper

extracted_from
Monitoring and Discovering Reward Hacking with Internal Representations during LLM Evaluations
(2026) · Leon Bergen · Usha Bhalla · Andrew Lee · Barak Widawsky +14

Neighborhood — ranked by edge-count