Unverified50% confidenceFactExact time
The alignment tax refers to the phenomenon where, when a model is overtrained to refuse harmful requests, its refusal boundary may generalize to normal harmless scenarios, resulting in excessive caution.
1
Sources
50%
Confidence
Long-term
Relevance
8/31/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedExcessive safety fine-tuning can cause models to over-refuse, undermining their practical utility.74% similarUnverifiedDrift detection involves a trade-off: setting the threshold too sensitively leads to frequent false alarms and unnecessary retraining, while setting it too loosely may allow model performance to continuously degrade without being detected.73% similarUnverified对齐税（Alignment Tax）指厂商在后训练阶段通过 RLHF 和安全微调降低有害输出概率时，往往导致模型在安全边界附近过度泛化，把无害请求也误判为危险请求并拒绝73% similarUnverifiedOver-refusal strategies reduce risk but harm model utility, especially in scenarios such as professional development and security research that require touching sensitive boundaries.72% similarVerifiedSome models carry the risk of over-optimization toward test sets, causing leaderboard scores to deviate from real-world performance (benchmark saturation).70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/829241API
curl https://kongchang.com/api/v1/knowledge/claims/829241MCP
get_claim(id=829241)