Unverified50% confidenceOpinionExact time
In LLM group decision-making, attribution labels (which agent proposed it) become a false consensus cue, making the model sensitive to social proof.
1
Sources
50%
Confidence
Long-term
Relevance
10/28/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn LLMs, the influence of human-like factors such as commonness, distinctiveness and characteristicness is transmitted to model ratings indirectly through 'factual truth' judgments.73% similarUnverifiedIn practice, the anchoring effect of pre-annotation results on humans must be considered: annotators may tend to accept the error types and severity levels given by the model, thereby reducing the independence and diversity of annotation.71% similarUnverifiedLLM evaluation is more objective than human evaluation, as it can score candidates consistently within a unified evaluation framework, avoiding the unfairness caused by cognitive biases such as fatigue, mood, and the primacy effect in human interviewers.70% similarUnverifiedHuman preference-driven evaluation methods compensate for the insufficient representativeness of traditional benchmarks (such as MMLU) in real-world usage scenarios.70% similarUnverifiedIn preference optimization, spurious correlation specifically refers to the non-causal association between input features and preference labels learned by the model, whose root cause lies in the observational bias of preference data.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1105924API
curl https://kongchang.com/api/v1/knowledge/claims/1105924MCP
get_claim(id=1105924)