Unverified50% confidenceEventExact time
In 2014, Stanford University's GloVe combined the advantages of matrix factorization methods and local context window methods by decomposing a global co-occurrence matrix.
1
Sources
50%
Confidence
Long-term
Relevance
9/8/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedGGUF now supports Tensor Parallelism, delivering approximately +30% improvement in inference throughput.57% similarUnverifiedLevy和Goldberg在2014年证明，Word2Vec的Skip-gram配合负采样实际上隐式地在分解一个PMI矩阵55% similarUnverifiedGPTQ is a layer-wise quantization method based on approximate second-order information, proposed by Frantar et al. in 2022, which minimizes quantization reconstruction error through the approximate inverse of the Hessian matrix.54% similarUnverifiedDE-9IM（维度扩展九交集模型）由Egenhofer和Herring于1991年提出，通过考察两个几何体的内部、边界、外部之间的九种两两交集，用3×3矩阵描述空间拓扑关系53% similarUnverifiedClosco claims its application scope covers multiple disciplines including computational materials science, molecular docking, deep learning ablation experiments, CFD simulation, and GWAS analysis.52% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/877244API
curl https://kongchang.com/api/v1/knowledge/claims/877244MCP
get_claim(id=877244)