Dnotitia Unveils STAR-KV, Achieving UP to 20x KV Cache Compression, Selected as an ICML 2026 Spotlight Paper

AI Summary

Dnotitia introduced STAR-KV, an advanced KV cache compression technology that can achieve up to 20x compression, significantly accelerating long-context AI attention computations and generation throughput. This innovation was selected as a Spotlight paper at ICML 2026, underscoring its importance in AI research.

Introduces a low-rank-based approach to KV cache compression, one of the key bottlenecks in long-context AI Speeds up attention computation by up to 6.9x and overall generation throughput by up to 3.1x, moving beyond memory savings to faster inference Selected as a Spotlight paper at ICML 2026, representing about 2.2% of reviewed submissions and about […]