[KongchangAI]
Unverified50% confidenceFactExact time

Across the three post-training paradigms of RLVR, SFT, and on-policy distillation, zeroing out less than 0.05% of the weights in the pretrained model matches the performance of full fine-tuning at scale.

1
Sources
50%
Confidence
Long-term
Relevance
11/10/2026
First Seen

Sources

Related Entities

Related Claims

Cite This Claim

Stable URI
https://kongchang.com/claim/1153748
API
curl https://kongchang.com/api/v1/knowledge/claims/1153748
MCP
get_claim(id=1153748)