arXiv:2609.02302 · v2
Improving Evaluation Realism with Inference-Time Compute and Deployment Scaffolds
摘要
A core obstacle to alignment evaluation is evaluation awareness: capable models can tell when they are being tested rather than deployed, weakening the conclusions a safety evaluation can support. We present two techniques that make simulated alignment evaluations harder to distinguish from real deployments. Our first technique, critique refinement, spends additional inference-time compute on each simulator action: the simulator generates multiple candidate actions, refines them using feedback from an instance of the target model on how to make them more realistic, and continues the evaluation with the most deployment-like candidate. Our second technique, DISH (Deployment-Imitating SWE-Agent Harness), wraps the target in an agent harness, reducing the gap between simulated and real deployment environments in coding settings. We test the techniques on multiple target models and find that applying both yields larger realism gains than either alone. We show that automated approaches can improve the realism of alignment evaluations, and that they compare favorably to spending the extra compute on more audit attempts.
作者备注：78 pages, 45 figures, 6 tables (8 figures in the main text). Accepted at NeurIPS 2026. Code: https://github.com/meridianlabs-ai/petri_dish and https://github.com/AxelAhlqvist1995/petri-bon ; reproduction assets: https://github.com/AxelAhlqvist1995/petri-realism-reproduction