Theory

Tag: Theory

2 items tagged with "Theory"

Quantize What Counts: More for Keys, Less for Values

July 5, 2026

ACL 2026 presentation on Quantize What Counts: More for Keys, Less for Values, explaining key-value norm disparity, key-prioritized quantization, and practical KV-cache compression guidance.

Poster

Quantize What Counts: More for Keys, Less for Values

Mohsen Hariri, Alan Luo, Weicong Chen, Tianyi Zhang, Qifan Wang, Xiaotian Han, Vipin Chaudhary

acl-2026

June 4, 2026

A geometry-driven mixed-precision KV-cache quantization poster showing that keys carry more information than values, so key-favored bit allocation preserves accuracy while reducing memory.