Sequential Attention: Making AI models leaner and faster without sacrificing accuracy

T1Covered by 1 source · first reported by Google — The Keyword / AI / Research / DeepMind / Developers at 04 Feb, 15:14 UTC

Algorithms & Theory

Sources

T1Sequential Attention: Making AI models leaner and faster without sacrificing accuracyGoogle — The Keyword / AI / Research / DeepMind / Developers04 Feb, 15:14

Sequential Attention: Making AI models leaner and faster without sacrificing accuracy · TechNews