Unverified50% confidenceFactExact time
In FM-Recon the foundation model parameters remain frozen, and only the lightweight feature refinement and fusion modules are trained, to avoid catastrophic forgetting and reduce computational overhead.
1
Sources
50%
Confidence
Long-term
Relevance
11/17/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIf the amount of fine-tuning data is limited, directly unfreezing all parameters for fine-tuning will almost inevitably trigger forgetting; a more reliable approach is to first freeze the backbone network and train only the final output head or adaptation layers.73% similarVerifiedThe 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.73% 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.70% similarUnverifiedAt moderate acceleration factors, FM-Recon achieved better reconstruction quality than baseline methods without foundation model features; at extremely high acceleration factors, the improvement was limited.70% similarUnverifiedThe approach used in this paper requires only lightweight temporal hints to guide the separation model to focus on time periods when calls occur, without requiring precise frame-by-frame masks, thus reducing dependence on detection accuracy and keeping the system detector-free and lightweight.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1179259API
curl https://kongchang.com/api/v1/knowledge/claims/1179259MCP
get_claim(id=1179259)