safety security
Shorter Reasoning Lost Faithfulness Before Monitorability
Three kinds of length pressure affected whether chains of thought reflected decisions more than whether they exposed interventions.
Summary
Three kinds of length pressure affected whether chains of thought reflected decisions more than whether they exposed interventions.
Researchers fine-tuned several models with fixed budgets, per-example length targets and group-relative length rewards. Chain-of-thought faithfulness fell in most settings, largely because answers became less consistent across related inputs. Monitorability was more robust: even shorter traces often still acknowledged when an input intervention changed the answer. The result varies by task and efficiency method.
Why it matters
Three kinds of length pressure affected whether chains of thought reflected decisions more than whether they exposed interventions.
Limits and context
No additional limitation was separately recorded.
Key claims
Three kinds of length pressure affected whether chains of thought reflected decisions more than whether they exposed interventions.
Evidence: source-2026-10-05-012
Sources
- arXiv preprint 2610.03509arXiv · primary research
Corrections
No corrections have been recorded for this story.