Speculate Deep and Accurate: Lossless and Training-Free Acceleration for Offloaded LLMs via Substitute Speculative Decoding

• URL: http://arxiv.org/abs/2509.18344v2
• Date: Wed, 08 Oct 2025 18:16:04 GMT
• Title: Speculate Deep and Accurate: Lossless and Training-Free Acceleration for Offloaded LLMs via Substitute Speculative Decoding
• Abstract summary: SubSpec is a plug-and-play method to accelerate parameter offloading.<n>It achieves 9.1x speedup for Qwen2.5 7B on MT-Bench and 12.5x speedup for Qwen2.5 32B on popular generation benchmarks.
• Score: 10.223767541511991
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: The immense model sizes of large language models (LLMs) challenge deployment on memory-limited consumer GPUs. Although model compression and parameter offloading are common strategies to address memory limitations, compression can degrade quality, and offloading maintains quality but suffers from slow inference. Speculative decoding presents a promising avenue to accelerate parameter offloading, utilizing a fast draft model to propose multiple draft tokens, which are then verified by the target LLM in parallel with a single forward pass. This method reduces the time-consuming data transfers in forward passes that involve offloaded weight transfers. Existing methods often rely on pretrained weights of the same family, but require additional training to align with custom-trained models. Moreover, approaches that involve draft model training usually yield only modest speedups. This limitation arises from insufficient alignment with the target model, preventing higher token acceptance lengths. To address these challenges and achieve greater speedups, we propose SubSpec, a plug-and-play method to accelerate parameter offloading that is lossless and training-free. SubSpec constructs a highly aligned draft model by generating low-bit quantized substitute layers from offloaded target LLM portions. Additionally, our method shares the remaining GPU-resident layers and the KV-Cache, further reducing memory overhead and enhance alignment. SubSpec achieves a high average acceptance length, delivering 9.1x speedup for Qwen2.5 7B on MT-Bench (8GB VRAM limit) and an average of 12.5x speedup for Qwen2.5 32B on popular generation benchmarks (24GB VRAM limit).

Related papers

• Unlocking Lossless Speedups in LLMs via Discrete Diffusion [67.4336730000437]
  Large Language Models (LLMs) owe much of their success to next-token prediction (NTP)<n>We introduce diffusion-augmented LLMs, a new class of models that defines an AR model distribution.<n>We decouple the parameters of these models into two sets: AR weights, trained using the standard NTP objective, and lightweight diffusion weights, trained to generate multiple tokens simultaneously.
  arXiv  Detail & Related papers  (2026-09-03T15:48:43Z)
• dParallel: Learnable Parallel Decoding for dLLMs [77.24184219948337]
  Diffusion large language models (dLLMs) offer parallel token prediction and lower inference latency.<n>Existing open-source models still require nearly token-length decoding steps to ensure performance.<n>We introduce dParallel, a simple and effective method that unlocks the inherent parallelism of dLLMs for fast sampling.
  arXiv  Detail & Related papers  (2025-09-30T16:32:52Z)
• SpecVLM: Fast Speculative Decoding in Vision-Language Models [14.243294546325714]
  Speculative decoding is a powerful way to accelerate autoregressive large language models (LLMs)<n>We study speculative decoding for vision-language models (VLMs)<n>We introduce SpecVLM, a practical system that delivers 1.5--2.3x end-to-end speedups over full autoregressive inference.
  arXiv  Detail & Related papers  (2025-09-15T11:53:56Z)
• Drag-and-Drop LLMs: Zero-Shot Prompt-to-Weights [75.83625828306839]
  textbfDrag-and-Drop LLMs (textitDnD) eliminates per-task training by mapping a handful of unlabeled task prompts directly to LoRA weight updates.<n>A lightweight text encoder distills each prompt batch into condition embeddings, which are then transformed by a cascaded hyper-convolutional decoder into the full set of LoRA matrices.
  arXiv  Detail & Related papers  (2025-06-19T15:38:21Z)
• Accelerating Diffusion LLMs via Adaptive Parallel Decoding [50.9948753314669]
  We introduce adaptive parallel decoding (APD), a novel method that dynamically adjusts the number of tokens sampled in parallel.<n>APD provides markedly higher throughput with minimal quality degradations on downstream benchmarks.
  arXiv  Detail & Related papers  (2025-05-31T06:10:10Z)
• DVD-Quant: Data-free Video Diffusion Transformers Quantization [86.89313228225666]
  DVD-Quant is a novel Data-free quantization framework for Video DiTs.<n>It achieves an approximately 2$times$ speedup over full-precision baselines on advanced DiT models.<n> DVD-Quant is the first to enable W4A4 PTQ for Video DiTs without compromising video quality.
  arXiv  Detail & Related papers  (2025-05-24T11:56:02Z)
• QuantSpec: Self-Speculative Decoding with Hierarchical Quantized KV Cache [67.84112700032007]
  Large Language Models (LLMs) are increasingly being deployed on edge devices for long-context settings.<n>In these scenarios, the Key-Value ( KV) cache is the primary bottleneck in terms of both GPU memory and latency.<n>We propose a novel self-speculative decoding framework, QuantSpec, where the draft model shares the architecture of the target model but employs a hierarchical 4-bit quantized KV cache and 4-bit quantized weights for acceleration.
  arXiv  Detail & Related papers  (2025-02-05T20:43:48Z)
• Huff-LLM: End-to-End Lossless Compression for Efficient LLM Inference [19.59857352852377]
  Large language models (LLMs) have continued to rapidly increase in size.<n>This has exacerbated the difficulty in running state of the art LLMs on small, edge devices.<n>We propose Huff-LLM, a method that lets users store LLM weights in compressed format.
  arXiv  Detail & Related papers  (2025-02-02T21:23:42Z)
• LoRC: Low-Rank Compression for LLMs KV Cache with a Progressive Compression Strategy [59.1298692559785]
  Key-Value ( KV) cache is crucial component in serving transformer-based autoregressive large language models (LLMs) Existing approaches to mitigate this issue include: (1) efficient attention variants integrated in upcycling stages; (2) KV cache compression at test time; and (3) KV cache compression at test time. We propose a low-rank approximation of KV weight matrices, allowing plug-in integration with existing transformer-based LLMs without model retraining. Our method is designed to function without model tuning in upcycling stages or task-specific profiling in test stages.
  arXiv  Detail & Related papers  (2024-10-04T03:10:53Z)
• MARLIN: Mixed-Precision Auto-Regressive Parallel Inference on Large Language Models [58.3342517278868]
  This paper describes the design of Mixed-precision AutoRegressive LINear kernels. It shows that batchsizes up to 16-32 can be supported with close to maximum ($4times$) quantization speedup. MarLIN accomplishes this via a combination of techniques, such as asynchronous memory access, complex task scheduling and pipelining.
  arXiv  Detail & Related papers  (2024-08-21T16:10:41Z)
• EfficientQAT: Efficient Quantization-Aware Training for Large Language Models [50.525259103219256]
  quantization-aware training (QAT) offers a solution by reducing memory consumption through low-bit representations with minimal accuracy loss.<n>We propose Efficient Quantization-Aware Training (EfficientQAT), a more feasible QAT algorithm.<n> EfficientQAT involves two consecutive phases: Block-wise training of all parameters (Block-AP) and end-to-end training of quantization parameters (E2E-QP)
  arXiv  Detail & Related papers  (2024-07-10T17:53:30Z)
• Direct Alignment of Draft Model for Speculative Decoding with Chat-Fine-Tuned LLMs [11.245862832561176]
  Training a high-quality draft model is required to enable inference acceleration via speculative decoding. We train Llama 2 Chat Drafter 115M, a draft model for Llama 2 Chat 7B or larger, with only 1.64% of the original size. Our results show that Llama 2 Chat Drafter 115M with speculative decoding achieves up to 2.3 block efficiency and 2.4$times$ speed-up.
  arXiv  Detail & Related papers  (2024-02-29T19:55:06Z)
• eDKM: An Efficient and Accurate Train-time Weight Clustering for Large Language Models [19.502740996431452]
  Differentiable KMeans Clustering, or DKM, has shown the state-of-the-art trade-off between compression ratio and accuracy regression. We propose a memory-efficient DKM implementation, eDKM powered by novel techniques to reduce the memory footprint of DKM by orders of magnitudes.
  arXiv  Detail & Related papers  (2023-09-02T15:16:35Z)
• Compress, Then Prompt: Improving Accuracy-Efficiency Trade-off of LLM Inference with Transferable Prompt [96.24800696597707]
  We introduce a new perspective to optimize this trade-off by prompting compressed models. We propose a soft prompt learning method where we expose the compressed model to the prompt learning process. Our experimental analysis suggests our soft prompt strategy greatly improves the performance of the 8x compressed LLaMA-7B model.
  arXiv  Detail & Related papers  (2023-05-17T20:45:13Z)

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.