Unverified50% confidenceCautionExact time
The challenge with score normalization is that when a knowledge base contains no truly relevant documents, the highest score after normalization may still be pushed up to the same level as relevant results from other bases, creating an illusion of high scores for irrelevant results.
1
Sources
50%
Confidence
Long-term
Relevance
9/6/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedLocomo评测数据集存在局限性，不同Memory框架使用不同的judgment prompt导致分数缺乏可比性72% similarUnverifiedVector retrieval is inherently weak at multi-hop reasoning; when an answer requires synthesizing scattered information fragments from multiple documents, pure similarity ranking often fails to retrieve all necessary context.70% similarUnverified自回归解码存在结构性缺陷，早期输出偏离事实会导致后续内容在错误方向上越走越远，形成雪崩效应69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/864515API
curl https://kongchang.com/api/v1/knowledge/claims/864515MCP
get_claim(id=864515)