Unverified50% confidenceFactExact time
In the field of continual learning, methods such as elastic weight consolidation (EWC) use Fisher information as a proxy for parameter importance, imposing stronger update penalties on important parameters to mitigate catastrophic forgetting.
1
Sources
50%
Confidence
Long-term
Relevance
10/20/2026
First Seen
Sources
Related Entities
Related Claims
Unverified研究者已提出弹性权重巩固(EWC)、渐进式神经网络等缓解灾难性遗忘的方案,但在超大规模模型上的有效性仍是活跃研究方向72% similarUnverifiedAcademia has proposed methods such as EWC (Elastic Weight Consolidation) and progressive networks to mitigate catastrophic forgetting, but their application to large-scale language models remains in an early exploratory stage70% similarUnverifiedFisher Merging uses the Fisher information matrix to assign importance weights to different parameters, preferentially preserving parameters that contribute more to each model's performance.65% similarUnverifiedThe main technical approaches explored for writing memory into weights include continual fine-tuning, meta-learning, and parameter-efficient fine-tuning methods such as LoRA.64% similarUnverifiedWhether a model can accurately locate key information in an ultra-long context and avoid the lost-in-the-middle phenomenon is a decisive factor in determining coding performance.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1073997API
curl https://kongchang.com/api/v1/knowledge/claims/1073997MCP
get_claim(id=1073997)