# Learning to Draft: Adaptive Speculative Decoding with Reinforcement Learning

arXiv: https://arxiv.org/abs/2603.01639 · HTML: https://mlsanity.com/paper/2603.01639/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Speculative decoding, one of 12 directions in Efficient LLM inference and serving.
- First posted on arXiv 2 Mar 2026.

## Placed in

- Efficient LLM inference and serving / [Speculative decoding](https://mlsanity.com/inference/speculative-decoding/)
- Deep reinforcement learning methods / [RL training of language and agent models](https://mlsanity.com/deeprl/rl-training-of-language-and-agent-models/)

## Built on by (most cited first, 5 of 5)

- [AngelSpec: Towards Real-World High Performance Inference with Speculative Decoding](https://mlsanity.com/paper/2607.25852/) (2026)
- [LibraSpec: Dynamic Diffusion-Based Speculative Decoding via Marginal-Gain-Driven Optimization](https://mlsanity.com/paper/2608.08721/) (2026)
- [Reciprocal Guidance: Orchestrating Draft and Verify Budgets for Advancing the Diffusion-AR Self-Speculation Frontier](https://mlsanity.com/paper/2609.34388/) (2026)
- [AS2D: Accelerating On-Demand Audio Understanding on Mobile Devices](https://mlsanity.com/paper/2609.37617/) (2026)
- [SP2ec: Adaptive Self-Speculative Decoding for Vision-Language Models](https://mlsanity.com/paper/2610.18735/) (2026)

## Abstract (arXiv)

Speculative decoding accelerates large language model (LLM) inference by using a small draft model to generate candidate tokens for a larger target model to verify. The efficacy of this technique hinges on the trade-off between the time spent on drafting candidates and verifying them. However, current state-of-the-art methods rely on a static time allocation, while recent dynamic approaches optimize for proxy metrics like acceptance length, often neglecting the true time cost and treating the drafting and verification phases in isolation. To address these limitations, we introduce Learning to Draft (LTD), a novel method that directly optimizes for throughput of each draft-and-verify cycle. We formulate the problem as a reinforcement learning environment and train two co-adaptive policies to dynamically coordinate the draft and verification phases. This encourages the policies to adapt to each other and explicitly maximize decoding efficiency. We conducted extensive evaluations on five diverse LLMs and four distinct tasks. Our results show that LTD achieves speedup ratios ranging from 2.24x to 4.32x, outperforming the state-of-the-art method Eagle3 up to 36.4%.