Unverified50% confidenceFactExact time
Multiplying the unlearning loss of the forget set by 0.0 causes the forget items to produce no gradients, meaning the model performs no real unlearning updates.
1
Sources
50%
Confidence
Long-term
Relevance
10/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedZero retention does not equal zero learning—the fact that data is not stored does not mean that patterns and information extracted from it will not influence the model in some form71% similarUnverified现有向量数据库记忆方案缺乏时效性衰减机制，无法区分记忆是否被采纳并产生正向效果，也没有机制合并冗余或删除有害记忆69% similarUnverifiedWhen semantically similar but inaccurate content is retrieved, the model may output incorrect recollections with high confidence, with no warning mechanism at the system level — this is the false memory problem.68% similarUnverifiedReplacing the forget set labels with zero causes the model to learn to predict those samples as class 0, and modifying labels is a different thing from eliminating the influence of samples.66% similarUnverifiedInterleaved training can prevent catastrophic forgetting, stopping the model from forgetting previously learned data when learning new types of data.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/998822API
curl https://kongchang.com/api/v1/knowledge/claims/998822MCP
get_claim(id=998822)