Expired50% confidenceFactTime unknown
RLHF training causes models to develop 'sycophancy,' meaning the model would rather fabricate information than admit ignorance.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
Related Claims
Unverified谄媚性（sycophancy）问题最早在RLHF训练范式中被系统性地识别出来82% similarUnverifiedA 2023 paper by OpenAI researchers found that models trained with RLHF tend to express agreement even when faced with clearly incorrect views (sycophancy bias).79% similarUnverifiedRLHF is essentially a behavioral-level constraint that does not alter the model's underlying capabilities; models trained with RLHF still retain latent knowledge of the restricted behaviors.78% similarUnverified模型在未收到明确指令时输出图片，意味着系统提示词设计或RLHF训练中存在激励偏差78% similarUnverifiedRLHF的人类介入发生在训练阶段并固化为模型权重，属于隐式编码，而表征工程干预在推理时操控内部表征，具有即时性和可逆性77% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/58587API
curl https://kongchang.com/api/v1/knowledge/claims/58587MCP
get_claim(id=58587)