claim
active
claim:unfaithfulness-is-likely-not-driven-by-a-preference-for-brevityUnfaithfulness is likely not driven by a preference for brevity
Unfaithful CoTs are longer.
Source paper
extracted_from(2025) · Yanda Chen · Joe Benton · Ansh Radhakrishnan · Jonathan Uesato +11
Neighborhood — ranked by edge-count
Papers (1)
paper
Findings (2)
finding
- Unfaithful CoTs are longer.
- Unfaithful CoTs are longer.
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.
- Interpretation of §3.
- Tested and rejected.
- Graph-predicted indirect effects that do not exist in the underlying model.
- Authors' interpretation of prompt variation results showing alignment faking disappears only when conflicting objective is removed
- Stakes of the problem.
- Why does the model use backwards reasoning in the unfaithful case, and why does it attend to the human's answer hint?question0.749Missing mechanistic piece.
- Practical implication.
- Takeaway of toy experiments.