AI Paper Atlas

Research map / Large language models / Reasoning and chain of thought

Re-SpS: A Reinforcement Learning Approach to Speculative Sampling

Chenan Wang, Daniel H. Shi, Haipeng Chen

AAAI 2026 · Large language models › Reasoning and chain of thought · approach: Speculative decoding · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes Re-SpS, a reinforcement learning approach that dynamically adjusts speculative sampling tree-structure hyperparameters (instead of static values used by methods like EAGLE-3) to reduce LLM inference latency.

Problem

Inference latency of LLMs; state-of-the-art tree-based speculative sampling (e.g., EAGLE-3) uses static tree-structure hyperparameters that limit flexibility and efficiency across diverse contexts and decoding settings.

Approach

Replace static tree-structure hyperparameters of speculative sampling with a learned RL policy that adapts them per context, so the drafting tree adjusts to each situation.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace SentenceBERT text embedding with reused draft-model hidden-state features as RL state (ablation)Re-SpS (standard PPO) · Avg speedup vs EAGLE-3, LLaMA 3.1-8Bspeedup: 1.044x -> 1.049xImproved
use feature vector instead of text embedding (ablation)Re-SpS (standard PPO) · Avg speedup vs EAGLE-3, Vicuna-13Bspeedup: 1.006x -> 1.028xImproved
increase action cache interval from 1 to 50 steps (ablation)Re-SpS · LLaMA 3.1-8B (80 questions)latency and speedImproved
max-entropy PPO instead of standard PPO (ablation)Re-SpS · LLaMA 3.1-8Bspeedup vs EAGLE-3: 1.049x -> 1.025xHurt
max-entropy PPO instead of standard PPO (ablation)Re-SpS · Vicuna-13Bspeedup vs EAGLE-3: 1.028x -> 1.033xImproved
replace static draft-tree hyperparameters with RL policyEAGLE-3 · LLaMA 3.3-70B mean over 5 benchmarksspeedup: 4.46× -> 4.88×Improved
replace static draft-tree hyperparameters with RL policyEAGLE-3 · LLaMA 3.1-8B HumanEvalspeedup: 3.65× -> 3.89×Improved
replace static hyperparameters with RL policyEAGLE-3 · CNN/DM (LLaMA 3.1-8B)speedup: 2.96× -> 2.87×Hurt
replace static hyperparameters with RL policyEAGLE-3 · MT-Bench (LLaMA 3.1-8B)speedup: 3.39× -> 3.43×No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Experimental setup

Baselines: EAGLE-3

Builds on: EAGLE-3 speculative sampling

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy