BASS: Batched Attention-optimized Speculative Sampling.
Haifeng QianSujan Kumar GonugondlaSungsoo HaMingyue ShangSanjay Krishna GoudaRamesh NallapatiSudipta SenguptaXiaofei MaAnoop DeorasPublished in: CoRR (2024)
Keyphrases
- monte carlo
- random sampling
- sample size
- neural network
- information systems
- sampling algorithm
- case study
- focus of attention
- visual attention
- sampling methods
- real time
- parameter estimation
- probability distribution
- multiresolution
- evolutionary algorithm
- high dimensional
- expert systems
- object recognition
- multi agent systems
- reinforcement learning
- knowledge base
- artificial intelligence