Unverified50% confidenceFactExact time
Candle's fused kernels silently produce zero gradients, causing certain parameters to not be updated during training at all.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/23/2026
First Seen
Valid until: 12/22/2026
Sources
Related Entities
Related Claims
VerifiedThe core idea of PEFT is to not update all parameters, but instead introduce a small number of trainable adapter parameters (such as LoRA's low-rank matrix decomposition), reducing VRAM and computation requirements while retaining performance close to full fine-tuning.60% similarUnverifieddegree参数只对poly核有效，gamma对linear核是无效参数，对这些组合遍历会产生无意义的重复训练57% similarVerifiedDuring LoRA fine-tuning, the original model parameters are frozen and only the inserted low-rank matrices are updated, significantly reducing VRAM requirements and training costs.57% similarUnverifiedBurn's kernel fusion path segfaults mid-training at the scale of billions of parameters.56% similarUnverifiedNeRF训练耗时长、推理速度慢，输出的隐式表示无法直接转化为可编辑的几何网格56% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/942817API
curl https://kongchang.com/api/v1/knowledge/claims/942817MCP
get_claim(id=942817)