Unverified50% confidenceFactExact time
The team used the identified tokens to construct a loss function and ran LoRA SFT, successfully reducing reasoning length but causing accuracy to drop.
1
Sources
50%
Confidence
Long-term
Relevance
9/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedA reasonable prediction loss combined with 0% downstream control accuracy is highly consistent with representation collapse, meaning the encoder has fooled the self-supervised objective without learning any useful dynamic information.68% similarUnverifiedThrough post-training methods, Swift-Qwen3.8-27B reduces thinking tokens by 58.3% and improves inference speed by 1.95×, while keeping accuracy loss within 1%.66% similarUnverifiedSFT fine-tuning is effective but causes a decline in general knowledge in small models—i.e., the phenomenon of catastrophic forgetting—due to the prominent capacity bottleneck of small models.65% similarVerifiedAfter geometric transformation, the transformation parameters must be recorded and the predicted boxes must be inverse-transformed to align them; ignoring this step will cause semantic misalignment in the consistency loss and harm training effectiveness65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/920995API
curl https://kongchang.com/api/v1/knowledge/claims/920995MCP
get_claim(id=920995)