# Make Every Draft Count: Hidden State based Speculative Decoding

arXiv: https://arxiv.org/abs/2602.21224 · HTML: https://mlsanity.com/paper/2602.21224/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Speculative decoding, one of 12 directions in Efficient LLM inference and serving.
- Builds on 26 of its 40 references that are in the map.
- First posted on arXiv 2 Feb 2026.

## Placed in

- Efficient LLM inference and serving / [Speculative decoding](https://mlsanity.com/inference/speculative-decoding/)

## Builds on (most cited first, 26 of 26)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Fast Inference from Transformers via Speculative Decoding](https://mlsanity.com/paper/2211.17192/) (2022)
- [SGLang: Efficient Execution of Structured Language Model Programs](https://mlsanity.com/paper/2312.07104/) (2023)
- [Accelerating Large Language Model Decoding with Speculative Sampling](https://mlsanity.com/paper/2302.01318/) (2023)
- [Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads](https://mlsanity.com/paper/2401.10774/) (2024)
- [Large Language Models](https://mlsanity.com/paper/2307.05782/) (2023)
- [EAGLE: Speculative Sampling Requires Rethinking Feature Uncertainty](https://mlsanity.com/paper/2401.15077/) (2024)
- [EAGLE-2: Faster Inference of Language Models with Dynamic Draft Trees](https://mlsanity.com/paper/2406.16858/) (2024)
- [SpecInfer: Accelerating Generative Large Language Model Serving with Tree-based Speculative Inference and Verification](https://mlsanity.com/paper/2305.09781/) (2023)
- [FlashInfer: Efficient and Customizable Attention Engine for LLM Inference Serving](https://mlsanity.com/paper/2501.01005/) (2025)
- [TriForce: Lossless Acceleration of Long Sequence Generation with Hierarchical Speculative Decoding](https://mlsanity.com/paper/2404.11912/) (2024)
- [Cascade Speculative Drafting for Even Faster LLM Inference](https://mlsanity.com/paper/2312.11462/) (2023)
- [Learning Harmonized Representations for Speculative Sampling](https://mlsanity.com/paper/2408.15766/) (2024)
- [Exploring Human-Like Translation Strategy with Large Language Models](https://mlsanity.com/paper/2305.04118/) (2023)
- [Lookahead: An Inference Acceleration Framework for Large Language Model with Lossless Generation Accuracy](https://mlsanity.com/paper/2312.12728/) (2023)
- [Speculative Decoding via Early-exiting for Faster LLM Inference with Thompson Sampling Control Mechanism](https://mlsanity.com/paper/2406.03853/) (2024)
- [SpecEE: Accelerating Large Language Model Inference with Speculative Early Exiting](https://mlsanity.com/paper/2504.08850/) (2025)
- [GRIFFIN: Effective Token Alignment for Faster Speculative Decoding](https://mlsanity.com/paper/2502.11018/) (2025)
- [SLED: A Speculative LLM Decoding Framework for Efficient Edge Serving](https://mlsanity.com/paper/2506.09397/) (2025)
- [Towards Fast Multilingual LLM Inference: Speculative Decoding and Specialized Drafters](https://mlsanity.com/paper/2406.16758/) (2024)
- [STree: Speculative Tree Decoding for Hybrid State-Space Models](https://mlsanity.com/paper/2505.14969/) (2025)
- [SwiftSpec: Ultra-Low Latency LLM Decoding by Scaling Asynchronous Speculative Decoding](https://mlsanity.com/paper/2506.11309/) (2025)
- [Efficient Speculative Decoding for Llama at Scale: Challenges and Solutions](https://mlsanity.com/paper/2508.08192/) (2025)
- [Jakiro: Boosting Speculative Decoding with Decoupled Multi-Head via MoE](https://mlsanity.com/paper/2502.06282/) (2025)
- [TPP-SD: Accelerating Transformer Point Process Sampling with Speculative Decoding](https://mlsanity.com/paper/2507.09252/) (2025)

## Built on by (most cited first, 4 of 4)

- [ReTrace: Rejected-Trajectory Conditioning for Speculative Decoding](https://mlsanity.com/paper/2608.29748/) (2026)
- [DFlow: Enabling Verifier Information Flow in Block Diffusion Speculative Decoding](https://mlsanity.com/paper/2609.06498/) (2026)
- [Carryover Drafting: Recycling Rejected States for Speculative Decoding](https://mlsanity.com/paper/2609.14717/) (2026)
- [Enabling Performant and Flexible Model-Internal Observability for LLM Inference](https://mlsanity.com/paper/2605.11093/) (2026)

## Abstract (arXiv)

Speculative decoding has emerged as a pivotal technique to accelerate LLM inference by employing a lightweight draft model to generate candidate tokens that are subsequently verified by the target model in parallel. However, while this paradigm successfully increases the arithmetic intensity of memory-bound inference, it causes significant compute inefficiency: the majority of draft tokens fail verification and are discarded, resulting in waste of computation. Motivated by the goal of recollecting this wasted computation, we propose a novel system that transforms discarded drafts into reusable tokens. Our key insight is to perform auto-regressive prediction at the hidden states level and postpone the integrating token information after the hidden states generation, so the draft hidden states are not contaminated by incorrect tokens, enabling hidden state reuse. To implement such a system, first we introduce a draft model architecture based on auto-regressive hidden states, which preserves richer semantics than token-based drafters to facilitate draft repurposing. Second, we design an efficient token information injection mechanism that leverages our specialized draft model to construct high-quality draft token trees and enables resampling tokens from verification failures. Third, we eliminate the overhead hidden in our design to further maximize hardware utilization. We conducted extensive evaluations against various baselines, demonstrating up to a 3.3x speedup against standard speculative decoding.