Unverified50% confidenceFactExact time

embedding模型的版本更新可能导致语义向量空间发生偏移，使原本高排名的相关文档被不太相关的文档取代

1

Sources

50%

Confidence

Long-term

Relevance

8/21/2026

First Seen

Sources

A Practical Guide to Building and Maintaining AI Eval Sets: Making LLM Application Quality Quantifiable

Related Claims

UnverifiedIf only the data of newly added classes is used to fine-tune an expanded classification head, the model is highly prone to catastrophic forgetting, rapidly losing its ability to recognize the original classes.66% similar Unverified由于历史记录同样受影响，模型名称消失更可能是前端统一显示逻辑层的问题，而非底层数据丢失64% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/780978

API

curl https://kongchang.com/api/v1/knowledge/claims/780978

MCP

get_claim(id=780978)