concept
active
concept:measuring-faithfulness-in-chain-of-thought-reasoning-lanham-et-al-2023Measuring Faithfulness in Chain-of-Thought Reasoning (Lanham et al. 2023)
Cited regarding possibility of encoding misaligned reasoning in benign chains-of-thought
Neighborhood — ranked by edge-count
Papers (2)
paper
- Reasoning Models Don't Always Say What They Thinkcitesmentions
Concepts (1)
concept
- Chain-of-thought faithfulnessrelated_toWhether written reasoning reflects actual model mechanisms.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Central conclusion.
- Medium through which eval awareness is often verbalized; target of intervention.
- Addressed partially in §3.3.4 but remains open especially for no-CoT settings
- Gap in current evaluation methods; current work relies on CoT monitoring which may miss unverbalized beliefs.
- Generality of methods.
- Faithfulness evaluation on reasoning-intensive, agentic and tool-use tasks remains unexploredclaim0.793Stated limitation/future work.
- Finding from Study 2 showing reasoning models remain vulnerable under both prompting and activation steering
- under what conditions does chain-of-thought reflect genuine uncertainty resolution versus a learned performance?question0.786Key question addressed by the task difficulty analysis comparing MMLU and GPQA-Diamond