Unverified50% confidenceFactExact time
RLHF中人类评估者在主观创作判断时倾向给予结构完整、措辞礼貌的答案更高分，这种偏差被称为奖励黑客的一种变体
1
Sources
50%
Confidence
Long-term
Relevance
7/6/2026
First Seen
Sources
The Anti-Slop Fiction Contest: How to Make AI Write Like It Isn't AI
hackernewshackernews7/4/2026
Related Claims
UnverifiedThe root cause of sycophantic behavior can be traced to the RLHF training paradigm, in which human evaluators prefer responses that agree with their own views.81% similarUnverifiedDuring the RLHF stage, human annotators tend to give higher ratings to responses that sound friendlier and more agreeable to the user, causing models to develop a systematic sycophantic tendency.80% similarVerifiedRLHF has limitations: human annotators have biases and inconsistencies, reward models may be over-optimized (reward hacking), and the process is costly and difficult to scale.78% similarUnverifiedResearch shows that human evaluators in RLHF tend to prefer longer, more detailed, and friendlier-toned responses, even when those responses are not superior in factual accuracy.78% similarUnverifiedRLHF的价值判断来自人类标注者的主观偏好，而CAI更依赖规则驱动的自动化批判机制78% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/114664API
curl https://kongchang.com/api/v1/knowledge/claims/114664MCP
get_claim(id=114664)