-
@ LessWrong (RSS Feed)
2025-05-24 19:17:55
It's really hard to make scheming evals look realistic
Published on May 24, 2025 7:17 PM GMTAbstractClaude 3.7 Sonnet easily detects when it's being evaluated for scheming. Surface‑level edits to evaluation scenarios, such as lengthening the prompts, or making conflict of objectives less salient, do improve realism of evaluation scenarios for LLMs, yet these improvements remain modest. The findings confirm that truly disguising an evaluation context demands removal of deep stylistic and structural cues rather than superficial prompt adjustments.For future LLMs the situation is likely to get worse as they will are likely to get better at evaluations awareness, which might become a big problem for detecting scheming via evaluations. Apollo Research published a https://arxiv.org/pdf/2412.04984
https://www.lesswrong.com/posts/TBk2dbWkg2F7dB3jb/it-s-really-hard-to-make-scheming-evals-look-realistic