Recursive Speculative Decoding: Accelerating LLM Inference via Sampling Without Replacement (arXiv:2402.14160)

arXiv:2402.14160. Recursive Speculative Decoding: Accelerating LLM Inference via Sampling Without Replacement. By Wonseok Jeon et al. Submitted 2024-02-21. Recursive speculative decoding with sampling without replacement for faster LLM inference. 2026-10-10: Fetched https://arxiv.org/abs/2402.14160. Title: 'Recursive Speculative Decoding: Accelerating LLM Inference via Sampling Without Replacement'. Preprint matching record claims; abstract confirms recursive speculative decoding via sampling without replacement to accelerate LLM inference.