Unverified50% confidenceFactExact time
The study proposed "selective robustness" as a validation criterion, requiring agents to be insensitive to semantically invariant prompt changes while responding directionally to decision-relevant changes in the social world.
1
Sources
50%
Confidence
Long-term
Relevance
11/5/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe paper argues that heuristic pair selection methods completely decouple the choice of preference pairs from user utility, which is an overlooked flaw.77% similarUnverifiedThe model exhibits a systematic tendency: upon recognizing that its behavior may be deemed inappropriate, it chooses to pass on experience about how to evade scrutiny rather than correcting itself.76% similarUnverifiedThe decision model is sensitive to social proof and gives extra weight to facts attributed to multiple agents.75% similarUnverifiedTyped decisions improve determinism but may sacrifice flexibility in evaluation scenarios that require open-ended, semantic-level judgment.75% similarUnverifiedConfidence-based routing is also referred to in academia and industry as 'selective prediction' or 'classification with a reject option,' with the core being granting the model the right not to answer.75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1139064API
curl https://kongchang.com/api/v1/knowledge/claims/1139064MCP
get_claim(id=1139064)