Shikhar Tuli

language model (3) mixture of expert (2) large language model (2) multi-head attention (2) state-space model (2) efficient computing (1) model pruning (1) posterior estimation (1) weight sharing (1) low-rank adaptation (1) inference optimization (1) inference acceleration (1) speculative decoding (1) multi-token prediction (1) non-autoregressive generation (1) flops reduction (1) neural network (1) hypothesis reduction (1) model compression (1) dynamic sampling (1)