Unverified50% confidenceFactExact time
In games between humans and agents, speaking first or accusing first was itself enough to change the course of the vote, and subsequent rebuttals were often ineffective.
1
Sources
50%
Confidence
Long-term
Relevance
10/28/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe probability of reviewers proactively raising their scores after a rebuttal is generally not high.71% similarUnverifiedThe authority bias effect nearly disappeared in multiple-choice pre-experiments, and the effect primarily appeared in free-form answer scenarios.70% similarUnverifiedWhen multiple models run in parallel, hallucinations often manifest as divergent answers rather than simple right or wrong, so they cannot be resolved by 'majority voting'.68% similarUnverified弓箭类武器上人类投票与客观数据分歧最大，因为人类会奖励「聪明的等待」行为而伤害数据无法体现68% similarUnverifiedResearch from 2023 showed that when multiple LLM agents engage in debate, if one agent expresses a strong position in early rounds, other agents tend to converge toward that position, potentially changing even correct original judgments—a phenomenon known as the herding effect.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1108767API
curl https://kongchang.com/api/v1/knowledge/claims/1108767MCP
get_claim(id=1108767)