Unverified50% confidenceFactExact time
ColBERT's core innovation is the MaxSim operation: for each token vector in a query, it finds the most similar token vector among all token vectors in the document, and sums these maximum similarity scores to obtain the final matching score.
1
Sources
50%
Confidence
Long-term
Relevance
10/10/2026
First Seen
Sources
Related Entities
Related Claims
Verified混合检索（Hybrid Search）将向量检索与BM25关键词检索的结果通过RRF（倒数排名融合）等算法合并，兼顾语义理解与精确匹配67% similarUnverifiedBeam search is a decoding algorithm commonly used in natural language generation that retains the k highest-probability candidate sequences at each step.66% similarUnverifiedVector search works by embedding text into a high-dimensional space and using similarity measures to find the segments closest to the query.66% similarUnverifiedIn the knowledge retrieval stage, the user's question is converted into a query vector via an embedding model, similarity is computed against the text chunk vectors in the knowledge base (typically using cosine similarity), and the top-K segments are returned by ranking.64% similarUnverifiedBeam search aims to generate the candidate sequence with the highest probability, whereas MBR decoding selects the candidate with the minimum expected translation error.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/993146API
curl https://kongchang.com/api/v1/knowledge/claims/993146MCP
get_claim(id=993146)