# Fast Large Language Model Collaborative Decoding via Speculation

arXiv: https://arxiv.org/abs/2502.01662 · HTML: https://mlsanity.com/paper/2502.01662/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 33 of its 37 references that are in the map.
- First posted on arXiv 1 Feb 2025.

## Builds on (most cited first, 30 of 33)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [OPT: Open Pre-trained Transformer Language Models](https://mlsanity.com/paper/2205.01068/) (2022)
- [Fast Inference from Transformers via Speculative Decoding](https://mlsanity.com/paper/2211.17192/) (2022)
- [Get To The Point: Summarization with Pointer-Generator Networks](https://mlsanity.com/paper/1704.04368/) (2017)
- [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)
- [Contrastive Decoding: Open-ended Text Generation as Optimization](https://mlsanity.com/paper/2210.15097/) (2022)
- [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)
- [Speculative Decoding: Exploiting Speculative Execution for Accelerating Seq2seq Generation](https://mlsanity.com/paper/2203.16487/) (2022)
- [DistillSpec: Improving Speculative Decoding via Knowledge Distillation](https://mlsanity.com/paper/2310.08461/) (2023)
- [TriForce: Lossless Acceleration of Long Sequence Generation with Hierarchical Speculative Decoding](https://mlsanity.com/paper/2404.11912/) (2024)
- [Decoding-Time Language Model Alignment with Multiple Objectives](https://mlsanity.com/paper/2406.18853/) (2024)
- [MLLM can see? Dynamic Correction Decoding for Hallucination Mitigation](https://mlsanity.com/paper/2410.11779/) (2024)
- [Contrastive Decoding Improves Reasoning in Large Language Models](https://mlsanity.com/paper/2309.09117/) (2023)
- [Decoding-time Realignment of Language Models](https://mlsanity.com/paper/2402.02992/) (2024)
- [Ensemble Learning for Heterogeneous Large Language Models with Deep Parallel Collaboration](https://mlsanity.com/paper/2404.12715/) (2024)
- [Judge Decoding: Faster Speculative Sampling Requires Going Beyond Model Alignment](https://mlsanity.com/paper/2501.19309/) (2025)
- [PaSS: Parallel Speculative Sampling](https://mlsanity.com/paper/2311.13581/) (2023)
- [Merge, Ensemble, and Cooperate! A Survey on Collaborative Strategies in the Era of Large Language Models](https://mlsanity.com/paper/2407.06089/) (2024)
- [Breaking the Ceiling of the LLM Community by Treating Token Generation as a Classification for Ensembling](https://mlsanity.com/paper/2406.12585/) (2024)
- [Determine-Then-Ensemble: Necessity of Top-k Union for Large Language Model Ensembling](https://mlsanity.com/paper/2410.03777/) (2024)
- [Purifying Large Language Models by Ensembling a Small Language Model](https://mlsanity.com/paper/2402.14845/) (2024)
- [Generation Meets Verification: Accelerating Large Language Model Inference with Smart Parallel Auto-Correct Decoding](https://mlsanity.com/paper/2402.11809/) (2024)
- [Optimized Multi-Token Joint Decoding with Auxiliary Model for LLM Inference](https://mlsanity.com/paper/2407.09722/) (2024)

## Built on by (most cited first, 12 of 12)

- [Harnessing Multiple Large Language Models: A Survey on LLM Ensemble](https://mlsanity.com/paper/2502.18036/) (2025)
- [Mitigating Visual Forgetting via Take-along Visual Conditioning for Multi-modal Long CoT Reasoning](https://mlsanity.com/paper/2503.13360/) (2025)
- [Small Drafts, Big Verdict: Information-Intensive Visual Reasoning via Speculation](https://mlsanity.com/paper/2510.20812/) (2025)
- [A Survey on Collaborating Small and Large Language Models for Performance, Cost-effectiveness, Cloud-edge Privacy, and Trustworthiness](https://mlsanity.com/paper/2510.13890/) (2025)
- [Collaborative Speculative Inference for Efficient LLM Inference Serving](https://mlsanity.com/paper/2503.10325/) (2025)
- [Overcoming Joint Intractability with Lossless Hierarchical Speculative Decoding](https://mlsanity.com/paper/2601.05724/) (2026)
- [When to Ensemble: Identifying Token-Level Points for Stable and Fast LLM Ensembling](https://mlsanity.com/paper/2510.15346/) (2025)
- [A Practical Investigation of Training-free Relaxed Speculative Decoding](https://mlsanity.com/paper/2607.08690/) (2026)
- [Universal Reasoner: A Single, Composable Plug-and-Play Reasoner for Frozen LLMs](https://mlsanity.com/paper/2505.19075/) (2025)
- [Rethinking LLM Ensembling from the Perspective of Mixture Models](https://mlsanity.com/paper/2605.00419/) (2026)
- [SpecFed: Accelerating Federated LLM Inference with Speculative Decoding and Compressed Transmission](https://mlsanity.com/paper/2604.25777/) (2026)
- [Revisiting Lossy Verification in Speculative Decoding: Mechanisms, Trade-offs, and Failure Modes](https://mlsanity.com/paper/2607.26627/) (2026)

## Abstract (arXiv)

Large Language Model (LLM) collaborative decoding techniques improve output quality by combining the outputs of multiple models at each generation step, but they incur high computational costs. In this paper, we introduce Collaborative decoding via Speculation (CoS), a novel framework that accelerates collaborative decoding without compromising performance. Inspired by Speculative Decoding--where a small proposal model generates tokens sequentially, and a larger target model verifies them in parallel, our approach builds on two key insights: (1) the verification distribution can be the combined distribution of both the proposal and target models, and (2) alternating each model as the proposer and verifier can further enhance efficiency. We generalize this method to collaboration among n models and theoretically prove that CoS is never slower than standard collaborative decoding, typically achieving faster speed. Extensive experiments demonstrate CoS is 1.11x-2.23x faster than standard collaborative decoding without compromising generation quality. Our code is available at https://github.com/Kamichanw/CoS/.