Unverified50% confidenceFactExact time
Vector quantization (VQ) maps a group of weights as a whole to the nearest codeword in a pre-learned codebook, storing only the codeword index.
1
Sources
50%
Confidence
Long-term
Relevance
10/8/2026
First Seen
Sources
Related Entities
Related Claims
Unverified自注意力通过将输入序列映射为查询Q、键K、值V三组向量，计算Q与K的点积相似度后经Softmax归一化得到注意力权重，再加权汇聚V向量70% similarUnverifiedQwen establishes dedicated lookup tables for phrase-level token combinations at the input stage through N-gram embeddings, concentrated in a large lookup layer at the front end of the model.67% similarUnverifiedIn QSA, a C4 block refers to every 4 adjacent tokens being organized into a single computational unit, with attention selection performed at the block level rather than the token level.65% similarUnverifiedThis RAG project extracts each FAQ as an independent Q&A pair as a complete semantic unit for vectorization, rather than using fixed-size chunking.64% similarUnverified倒排乘积量化（IVFPQ）在PQ基础上引入倒排索引结构，通过聚类将向量库分区，将检索复杂度从线性降低至亚线性64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/988394API
curl https://kongchang.com/api/v1/knowledge/claims/988394MCP
get_claim(id=988394)