FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness
Tri Dao, Stefano Ermon, Dan Fu, Christopher Ré, Atri Rudra
Topics & Concepts
Computer scienceAlgorithmArtificial intelligenceMathematicsNoise (video)Set (abstract data type)Sequence (biology)Feature (linguistics)Focus (optics)Stability (learning theory)Gaze Tracking and Assistive TechnologyPersonal Information Management and User BehaviorInteractive and Immersive Displays