Speculative Decoding in Decentralized LLM Inference: Turning Communication Latency into Computation Throughput

• URL: http://arxiv.org/abs/2511.11733v1
• Date: Thu, 13 Nov 2025 06:36:33 GMT
• Title: Speculative Decoding in Decentralized LLM Inference: Turning Communication Latency into Computation Throughput
• Authors: Jingwei Song, Wanyi Chen, Xinyuan Song, Max, Chris Tong, Gufeng Chen, Tianyi Zhao, Eric Yang, Bill Shi, Lynn Ai,
• Abstract summary: We present Decentralized Speculative Decoding (DSD), a plug-and-play framework for decentralized inference.<n>We introduce an adaptive speculative verification strategy that adjusts acceptance thresholds by token-level semantic importance, delivering an additional 15% to 20% end-to-end speedup without retraining.<n>DSD achieves up to 2.56x speedup on HumanEval and 2.59x on GSM8K, surpassing the Eagle3 baseline while preserving accuracy.
• Score: 8.480238305117298
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Speculative decoding accelerates large language model (LLM) inference by using a lightweight draft model to propose tokens that are later verified by a stronger target model. While effective in centralized systems, its behavior in decentralized settings, where network latency often dominates compute, remains under-characterized. We present Decentralized Speculative Decoding (DSD), a plug-and-play framework for decentralized inference that turns communication delay into useful computation by verifying multiple candidate tokens in parallel across distributed nodes. We further introduce an adaptive speculative verification strategy that adjusts acceptance thresholds by token-level semantic importance, delivering an additional 15% to 20% end-to-end speedup without retraining. In theory, DSD reduces cross-node communication cost by approximately (N-1)t1(k-1)/k, where t1 is per-link latency and k is the average number of tokens accepted per round. In practice, DSD achieves up to 2.56x speedup on HumanEval and 2.59x on GSM8K, surpassing the Eagle3 baseline while preserving accuracy. These results show that adapting speculative decoding for decentralized execution provides a system-level optimization that converts network stalls into throughput, enabling faster distributed LLM inference with no model retraining or architectural changes.

Related papers

• Orchestrating Multimodal DNN Workloads in Wireless Neural Processing [57.510786937781866]
  In edge inference, wireless resource allocation and accelerator deep neural computation (DNN) scheduling have yet to be co-optimized in an end-to-end manner.<n>This paper investigates a paradigm that integrates wireless transmission and multi-core execution into a unified end-to-end pipeline.
  arXiv  Detail & Related papers  (2026-03-02T17:25:43Z)
• Accelerating Decentralized Optimization via Overlapping Local Steps [6.713278402701195]
  We propose a novel approach to accelerate decentralized computation without sacrificing theoretical guarantees.<n>We show OLDSGD retains the same iteration as Local Decentralized SGDOLDD while improving per-iteration convergence under different levels of communication delays.
  arXiv  Detail & Related papers  (2026-01-04T11:40:22Z)
• Logit-Entropy Adaptive Stopping Heuristic for Efficient Chain-of-Thought Reasoning [0.0]
  Chain-of-Thought (CoT) prompting is a key technique for enabling complex reasoning in large language models.<n>We introduce LEASH: Logit-Entropy Adaptive Stopping Heuristic, a training-free decoding algorithm that adaptively halts rationale generation.
  arXiv  Detail & Related papers  (2025-11-06T18:43:16Z)
• FusionLLM: A Decentralized LLM Training System on Geo-distributed GPUs with Adaptive Compression [55.992528247880685]
  Decentralized training faces significant challenges regarding system design and efficiency. We present FusionLLM, a decentralized training system designed and implemented for training large deep neural networks (DNNs) We show that our system and method can achieve 1.45 - 9.39x speedup compared to baseline methods while ensuring convergence.
  arXiv  Detail & Related papers  (2024-10-16T16:13:19Z)
• Towards Communication-efficient Federated Learning via Sparse and Aligned Adaptive Optimization [90.08459757321405]
  Federated Adam (FedAdam) algorithms suffer from a threefold increase in uplink communication overhead.<n>We propose a novel sparse FedAdam algorithm called FedAdam-SSM, wherein distributed devices sparsify the updates local model parameters and moment estimates.<n>By minimizing the divergence bound between the model trained by FedAdam-SSM and centralized Adam, we optimize the SSM to mitigate the learning performance degradation caused by sparsification error.
  arXiv  Detail & Related papers  (2024-05-28T07:56:49Z)
• Lossless Acceleration of Large Language Model via Adaptive N-gram Parallel Decoding [2.642212767247493]
  We introduce Adaptive N-gram Parallel Decoding (ANPD), which accelerates inference by allowing the simultaneous generation of multiple tokens. ANPD preserves the integrity of the original output while enhancing processing speed. In experiments, models such as LLaMA and its fine-tuned variants have shown speed improvements up to 3.67x.
  arXiv  Detail & Related papers  (2024-04-10T16:11:09Z)
• Sparse Decentralized Federated Learning [35.32297764027417]
  Decentralized Federated Learning (DFL) enables collaborative model training without a central server but faces challenges in efficiency, stability, and trustworthiness.<n>We introduce a sparsity constraint on the shared model, leading to Sparse DFL (SDFL), and propose a novel algorithm, CEPS.<n> Numerical experiments validate the effectiveness of the proposed algorithm in improving communication and efficiency while maintaining a high level of trustworthiness.
  arXiv  Detail & Related papers  (2023-08-31T12:22:40Z)
• An Adaptive Device-Edge Co-Inference Framework Based on Soft Actor-Critic [72.35307086274912]
  High-dimension parameter model and large-scale mathematical calculation restrict execution efficiency, especially for Internet of Things (IoT) devices. We propose a new Deep Reinforcement Learning (DRL)-Soft Actor Critic for discrete (SAC-d), which generates the emphexit point, emphexit point, and emphcompressing bits by soft policy iterations. Based on the latency and accuracy aware reward design, such an computation can well adapt to the complex environment like dynamic wireless channel and arbitrary processing, and is capable of supporting the 5G URL
  arXiv  Detail & Related papers  (2022-01-09T09:31:50Z)
• Two-Timescale End-to-End Learning for Channel Acquisition and Hybrid Precoding [94.40747235081466]
  We propose an end-to-end deep learning-based joint transceiver design algorithm for millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) systems. We develop a DNN architecture that maps the received pilots into feedback bits at the receiver, and then further maps the feedback bits into the hybrid precoder at the transmitter.
  arXiv  Detail & Related papers  (2021-10-22T20:49:02Z)
• Multi-Exit Semantic Segmentation Networks [78.44441236864057]
  We propose a framework for converting state-of-the-art segmentation models to MESS networks. specially trained CNNs that employ parametrised early exits along their depth to save during inference on easier samples. We co-optimise the number, placement and architecture of the attached segmentation heads, along with the exit policy, to adapt to the device capabilities and application-specific requirements.
  arXiv  Detail & Related papers  (2021-06-07T11:37:03Z)
• Overlap Local-SGD: An Algorithmic Approach to Hide Communication Delays in Distributed SGD [32.03967072200476]
  We propose an algorithmic approach named OverlapLocal-Local-Local-SGD (Local momentum variant) We achieve this by adding an anchor model on each node. After multiple local updates, locally trained models will be pulled back towards the anchor model rather than communicating with others.
  arXiv  Detail & Related papers  (2020-02-21T20:33:49Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.