concept
active
concept:steganography-in-cotSteganography in CoT
Way models might evade CoT monitoring.
Neighborhood — ranked by edge-count
Papers (1)
paper
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.
- Encoding misaligned reasoning in seemingly benign chain-of-thought; possible future mechanism for alignment faking
- Covert encoding of information in NLA explanations beyond their overt natural language meaning.
- Information hidden in CoT phrasing not understandable by humans.
- Monitoring CoT to detect misaligned intentions; paper tests its reliability.
- Chain-of-thought prompting baseline used for comparison in creative writing and other tasks
- A VS variant that adds chain-of-thought reasoning before generating the distribution of responses with probabilities
- A two-stage framework that separates rationale generation and answer inference by incorporating vision and language modalities.
- From paraphrasing results.