# Speculative Decoding via Early-exiting for Faster LLM Inference with Thompson Sampling Control Mechanism

arXiv: https://arxiv.org/abs/2406.03853 · HTML: https://mlsanity.com/paper/2406.03853/

## Key facts

- Cited by 32 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 19 of its 30 references that are in the map.
- First posted on arXiv 6 Jun 2024.

## Placed in

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

## Builds on (most cited first, 19 of 19)

- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers](https://mlsanity.com/paper/2210.17323/) (2022)
- [Fast Inference from Transformers via Speculative Decoding](https://mlsanity.com/paper/2211.17192/) (2022)
- [SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models](https://mlsanity.com/paper/2211.10438/) (2022)
- [Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization](https://mlsanity.com/paper/1808.08745/) (2018)
- [SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shot](https://mlsanity.com/paper/2301.00774/) (2023)
- [Accelerating Large Language Model Decoding with Speculative Sampling](https://mlsanity.com/paper/2302.01318/) (2023)
- [BranchyNet: Fast Inference via Early Exiting from Deep Neural Networks](https://mlsanity.com/paper/1709.01686/) (2017)
- [Introduction to Multi-Armed Bandits](https://mlsanity.com/paper/1904.07272/) (2019)
- [Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning](https://mlsanity.com/paper/2310.06694/) (2023)
- [Depth-Adaptive Transformer](https://mlsanity.com/paper/1910.10073/) (2019)
- [Accelerating LLM Inference with Staged Speculative Decoding](https://mlsanity.com/paper/2308.04623/) (2023)
- [Symbolic Chain-of-Thought Distillation: Small Models Can Also "Think" Step-by-Step](https://mlsanity.com/paper/2306.14050/) (2023)
- [Fast and Robust Early-Exiting Framework for Autoregressive Language Models with Synchronized Parallel Decoding](https://mlsanity.com/paper/2310.05424/) (2023)
- [Lifting the Curse of Capacity Gap in Distilling Language Models](https://mlsanity.com/paper/2305.12129/) (2023)
- [PreQuant: A Task-agnostic Quantization Approach for Pre-trained Language Models](https://mlsanity.com/paper/2306.00014/) (2023)

## Built on by (most cited first, 30 of 32)

- [Judge Decoding: Faster Speculative Sampling Requires Going Beyond Model Alignment](https://mlsanity.com/paper/2501.19309/) (2025)
- [LongCat-Flash Technical Report](https://mlsanity.com/paper/2509.01322/) (2025)
- [A Survey on Parallel Text Generation: From Parallel Decoding to Diffusion Language Models](https://mlsanity.com/paper/2508.08712/) (2025)
- [Speculative Decoding and Beyond: An In-Depth Survey of Techniques](https://mlsanity.com/paper/2502.19732/) (2025)
- [DREAM: Drafting with Refined Target Features and Entropy-Adaptive Cross-Attention Fusion for Multimodal Speculative Decoding](https://mlsanity.com/paper/2505.19201/) (2025)
- [LogitSpec: Accelerating Retrieval-based Speculative Decoding via Next Next Token Speculation](https://mlsanity.com/paper/2507.01449/) (2025)
- [TALON: Confidence-Aware Speculative Decoding with Adaptive Token Trees](https://mlsanity.com/paper/2601.07353/) (2026)
- [Fuzzy Speculative Decoding for a Tunable Accuracy-Runtime Tradeoff](https://mlsanity.com/paper/2502.20704/) (2025)
- [Certified Self-Consistency: Statistical Guarantees and Test-Time Training for Reliable Reasoning in LLMs](https://mlsanity.com/paper/2510.17472/) (2025)
- [Guided Speculative Inference for Efficient Test-Time Alignment of LLMs](https://mlsanity.com/paper/2506.04118/) (2025)
- [Accelerating Large-Scale Reasoning Model Inference with Sparse Self-Speculative Decoding](https://mlsanity.com/paper/2512.01278/) (2025)
- [SPIN: Accelerating Large Language Model Inference with Heterogeneous Speculative Models](https://mlsanity.com/paper/2503.15921/) (2025)
- [AdaDecode: Accelerating LLM Decoding with Adaptive Layer Parallelism](https://mlsanity.com/paper/2506.03700/) (2025)
- [Constrained Decoding with Speculative Lookaheads](https://mlsanity.com/paper/2412.10418/) (2024)
- [Semi-Clairvoyant Scheduling of Speculative Decoding Requests to Minimize LLM Inference Latency](https://mlsanity.com/paper/2505.17074/) (2025)
- [CAS-Spec: Cascade Adaptive Self-Speculative Decoding for On-the-Fly Lossless Inference Acceleration of LLMs](https://mlsanity.com/paper/2510.26843/) (2025)
- [Fail Fast, Win Big: Rethinking the Drafting Strategy in Speculative Decoding via Diffusion LLMs](https://mlsanity.com/paper/2512.20573/) (2025)
- [Make Every Draft Count: Hidden State based Speculative Decoding](https://mlsanity.com/paper/2602.21224/) (2026)
- [Multi-Drafter Speculative Decoding with Alignment Feedback](https://mlsanity.com/paper/2604.05417/) (2026)
- [GPT-OSS-20B: A Comprehensive Deployment-Centric Analysis of OpenAI's Open-Weight Mixture of Experts Model](https://mlsanity.com/paper/2508.16700/) (2025)
- [AMBEDKAR-A Multi-level Bias Elimination through a Decoding Approach with Knowledge Augmentation for Robust Constitutional Alignment of Language Models](https://mlsanity.com/paper/2509.02133/) (2025)
- [Dynamic Rebatching for Efficient Early-Exit Inference with DREX](https://mlsanity.com/paper/2512.15705/) (2025)
- [OLIVIA: Online Learning via Inference-time Action Adaptation for Decision Making in LLM ReAct Agents](https://mlsanity.com/paper/2605.11169/) (2026)
- [COUNTDOWN: Contextually Sparse Activation Filtering Out Unnecessary Weights in Down Projection](https://mlsanity.com/paper/2505.17701/) (2025)
- [Hierarchical Verification of Speculative Beams for Accelerating LLM Inference](https://mlsanity.com/paper/2508.03726/) (2025)
- [Pipeline Parallelism is All You Need for Optimized Early-Exit Based Self-Speculative Decoding](https://mlsanity.com/paper/2509.19368/) (2025)
- [A Component-Based Survey of Interactions between Large Language Models and Multi-Armed Bandits](https://mlsanity.com/paper/2601.12945/) (2026)
- [The Diminishing Returns of Early-Exit Decoding in Modern LLMs](https://mlsanity.com/paper/2603.23701/) (2026)
- [FIRP: Faster LLM inference via future intermediate representation prediction](https://mlsanity.com/paper/2410.20488/) (2024)
- [AxLLM: accelerator architecture for large language models with computation reuse capability](https://mlsanity.com/paper/2509.22512/) (2025)

## Abstract (arXiv)

The recent advancements in large language models (LLMs) have been extraordinary, yet the escalating inference costs associated with them present challenges in real-world applications. To address these challenges, we propose a novel approach called Early-exiting Speculative Decoding (EESD) with lossless acceleration. Specifically, EESD utilizes a segment of the LLM to generate draft tokens, incorporating Early-exiting structures after the first N layers. To enhance the quality of draft tokens, a self-distillation method is integrated. This early-exiting design not only reduces deployment and training costs but also significantly accelerates the token generation speed. Moreover, we introduce a novel sampling mechanism that leverages Thompson Sampling to regulate the generation processes, automatically determining the quantity of draft tokens in each round. The original LLM is then employed to validate these draft tokens through a single forward pass, and thus guarantees that the final output text maintains a distribution consistent with vanilla auto-regressive decoding. The experimental results on both 13B and 70B models demonstrate that our approach decodes tokens at a markedly accelerated rate compared to prior methods, showing the effectiveness of our approach.