Anshumali Shrivastava

To Compress or Not? Pushing the Frontier of Lossless GenAI Model Weights Compression with Exponent Concentration

REFRAG: Rethinking RAG based Decoding

RACE Attention: A Strictly Linear-Time Attention for Long-Sequence Training

70% Size, 100% Accuracy: Lossless LLM Compression for Efficient GPU Inference via Dynamic-Length Float (DFloat11)

Sketch to Adapt: Fine-Tunable Sketches for Efficient LLM Adaptation

Breaking the Frozen Subspace: Importance Sampling for Low-Rank Optimization in LLM Pretraining

LeanQuant: Accurate and Scalable Large Language Model Quantization with Loss-error-aware Grid

SpaLLM: Unified Compressive Adaptation of Large Language Models with Sketching