← Search

Ruihang Lai

2 accepted papers

2025

MagicDec: Breaking the Latency-Throughput Tradeoff for Long Context Generation with Speculative Decoding

ICLR 2025poster

Large Language Models (LLMs) have become more prevalent in long-context applications such as interactive chatbots, document analysis, and agent workflows, but it is challenging to serve long-context requests with low latency and high throughput. Speculative decoding (SD) is a widely used technique t…

2022

Tensor Program Optimization with Probabilistic Programs

NeurIPS 2022accept

Automatic optimization for tensor programs becomes increasingly important as we deploy deep learning in various environments, and efficient optimization relies on a rich search space and effective search. Most existing efforts adopt a search space which lacks the ability to efficiently enable domain…

Cited by 36SourcePDFScholar