Unverified70% confidenceFactTime unknown
谄媚性（sycophancy）问题最早在RLHF训练范式中被系统性地识别出来
1
Sources
70%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedRLHF training causes models to develop 'sycophancy,' meaning the model would rather fabricate information than admit ignorance.82% similarUnverified早期对话AI系统普遍存在「奉承偏差」（Sycophancy）问题，在基于RLHF训练的模型中尤为突出79% 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% 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.77% 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).76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/30917API
curl https://kongchang.com/api/v1/knowledge/claims/30917MCP
get_claim(id=30917)