# Speculative decoding

Area: [Efficient LLM inference and serving](https://mlsanity.com/inference/index.md). HTML: https://mlsanity.com/inference/speculative-decoding/

## Key facts

- Speculative decoding: 522 papers on arXiv since 2021 (data through 13 Oct 2026).
- 465 of its 522 mapped papers (89%) cite “Fast Inference from Transformers via Speculative Decoding” (2022, arXiv 2211.17192).
- 382 of its 522 mapped papers (73%) cite “Accelerating Large Language Model Decoding with Speculative Sampling” (2023, arXiv 2302.01318).
- 30 new papers 24 Sep–8 Oct.

## Most cited here

- [Fast Inference from Transformers via Speculative Decoding](https://mlsanity.com/paper/2211.17192/) (2022), cited by 465 papers here
- [Accelerating Large Language Model Decoding with Speculative Sampling](https://mlsanity.com/paper/2302.01318/) (2023), cited by 382 papers here
- [Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads](https://mlsanity.com/paper/2401.10774/) (2024), cited by 353 papers here
- [EAGLE: Speculative Sampling Requires Rethinking Feature Uncertainty](https://mlsanity.com/paper/2401.15077/) (2024), cited by 296 papers here
- [EAGLE-2: Faster Inference of Language Models with Dynamic Draft Trees](https://mlsanity.com/paper/2406.16858/) (2024), cited by 224 papers here
- [SpecInfer: Accelerating Generative Large Language Model Serving with Tree-based Speculative Inference and Verification](https://mlsanity.com/paper/2305.09781/) (2023), cited by 210 papers here
- [EAGLE-3: Scaling up Inference Acceleration of Large Language Models via Training-Time Test](https://mlsanity.com/paper/2503.01840/) (2025), cited by 182 papers here
- [Break the Sequential Dependency of LLM Inference Using Lookahead Decoding](https://mlsanity.com/paper/2402.02057/) (2024), cited by 167 papers here
- [Unlocking Efficiency in Large Language Model Inference: A Comprehensive Survey of Speculative Decoding](https://mlsanity.com/paper/2401.07851/) (2024), cited by 163 papers here
- [REST: Retrieval-Based Speculative Decoding](https://mlsanity.com/paper/2311.08252/) (2023), cited by 112 papers here

## New (24 Sep–8 Oct)

- [Contextual Correction of Parallel Draftsfor Speculative Decoding](https://mlsanity.com/paper/2610.12570/) (2026-10-08)
- [Where Draft Trees Lose Target Mass: Exit-Guided Speculative Decoding](https://mlsanity.com/paper/2610.11750/) (2026-10-08)
- [Speedbumps: Rejection Attacks on Speculative Decoding](https://mlsanity.com/paper/2610.10929/) (2026-10-07)
- [Training Parallel Speculative Draft Models by Directly Minimizing Expected Decoding Rounds](https://mlsanity.com/paper/2610.10411/) (2026-10-07)
- [Secure Speculative Decoding for Large Language Models](https://mlsanity.com/paper/2610.08678/) (2026-10-06)
- [Nucleus Speculative Decoding: Plausibility-Aware Verification Beyond Exact Distribution](https://mlsanity.com/paper/2610.07822/) (2026-10-06)
- [APEX: Speculate smarter, not deeper](https://mlsanity.com/paper/2610.07780/) (2026-10-06)
- [DLoop: Looped Speculative Decoding](https://mlsanity.com/paper/2610.07659/) (2026-10-06)
- [Trajectory-Retrieval Speculative Decoding: When Does a Model's Own History Help?](https://mlsanity.com/paper/2610.07350/) (2026-10-05)
- [SchemaFill: Efficient LLM Tool Calling via Slot-Parallel Speculative Decoding](https://mlsanity.com/paper/2610.07086/) (2026-10-05)
- [D-Loop: Looped Diffusion Drafting for Speculative Decoding](https://mlsanity.com/paper/2610.06011/) (2026-10-05)
- [SharpDraft: Accelerating Long-Context Speculative Decoding with Cardinality-Aware Query Scaling](https://mlsanity.com/paper/2610.05106/) (2026-10-04)
- [BitNest: Bit-Nested Speculative Decoding for Memory-Efficient LLM Inference Acceleration](https://mlsanity.com/paper/2610.02800/) (2026-10-02)
- [DRelay: Global Draft Context for Prefix-Aware Parallel Speculative Decoding Repair](https://mlsanity.com/paper/2610.01439/) (2026-10-01)
- [AgSpec: Pushing the Limits of Retrieval-Based Speculative Decoding in Coding Agent Pipelines](https://mlsanity.com/paper/2610.01108/) (2026-10-01)
- [HAWK: Rethinking Multimodal Drafting for Speculative Decoding](https://mlsanity.com/paper/2610.00623/) (2026-09-30)
- [UBTree: Parallel Tree Drafting via Unigram and Bigram Models for Speculative Decoding](https://mlsanity.com/paper/2609.39972/) (2026-09-30)
- [Recovering Off-Policy Supervision for Speculative Decoding](https://mlsanity.com/paper/2609.38795/) (2026-09-30)
- [CAST: Cost-Aware Speculative Trees from One-Pass Block Drafters](https://mlsanity.com/paper/2610.00321/) (2026-09-29)
- [DEdit: Iterative Draft Editing for Speculative Decoding](https://mlsanity.com/paper/2609.38510/) (2026-09-29)
- [AS2D: Accelerating On-Demand Audio Understanding on Mobile Devices](https://mlsanity.com/paper/2609.37617/) (2026-09-29)
- [DScale: Scaling Block-Diffusion Speculative Decoding with Adaptive Verification](https://mlsanity.com/paper/2609.37532/) (2026-09-29)
- [LongSpark: Efficient speculative decoding with a fixed-cost parallel drafter](https://mlsanity.com/paper/2609.37029/) (2026-09-29)
- [SEED: Self-Speculative Decoding via Implicit Encoder-Decoder](https://mlsanity.com/paper/2609.36590/) (2026-09-29)
- [Draft in Parallel, Condition Through Depth: Adjacent Causal Injection for Speculative Decoding](https://mlsanity.com/paper/2609.36173/) (2026-09-28)
- [BV Loss: Block Verification-Aware Loss for Block Diffusion Speculative Decoding](https://mlsanity.com/paper/2609.34832/) (2026-09-28)
- [Reciprocal Guidance: Orchestrating Draft and Verify Budgets for Advancing the Diffusion-AR Self-Speculation Frontier](https://mlsanity.com/paper/2609.34388/) (2026-09-28)
- [Tsubame: Tree Replay for Diffusion-Based Speculative Decoding](https://mlsanity.com/paper/2609.33652/) (2026-09-27)
- [Acoustic Progress Propagation for Long-Horizon Speculative Decoding in ASR](https://mlsanity.com/paper/2609.33245/) (2026-09-27)
- [Mentored Decoding: Faster Inference meets Boosting](https://mlsanity.com/paper/2609.30474/) (2026-09-24)