Unverified50% confidenceFactExact time
When the teacher has non-uniform style preferences, SFT produces off-axis style drift along a non-zero task mean under population updates, which becomes the root cause of forgetting.
1
Sources
50%
Confidence
Long-term
Relevance
10/5/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe paper proves that starting from the same perfectly fitted checkpoint, SFT's forgetting over a finite training interval has a lower bound that is strictly greater than zero.66% similarUnverified作者将该现象解读为梯度下降在训练中刻意将某些模式和概念安放在噪声区间62% similarUnverified教师模型的不确定性在学生token序列上是可因式分解的，分歧被推迟到后续token才解决，可利用链式法则找回61% similarUnverifiedWhen the underlying KT model changes (e.g., migrating from BKT to DKT) or instructional priorities are adjusted, the mastery threshold must be recalibrated.61% similarUnverifiedSupervised fine-tuning (SFT) tends to cause more severe catastrophic forgetting than reinforcement fine-tuning (RFT).60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/975968API
curl https://kongchang.com/api/v1/knowledge/claims/975968MCP
get_claim(id=975968)