SJD-PAC: Accelerating Speculative Jacobi Decoding via Proactive Drafting and Adaptive Continuation

• URL: http://arxiv.org/abs/2603.18599v1
• Date: Thu, 19 Mar 2026 08:15:18 GMT
• Title: SJD-PAC: Accelerating Speculative Jacobi Decoding via Proactive Drafting and Adaptive Continuation
• Abstract summary: Speculative Jacobi Decoding (SJD) offers a draft-model-free approach to accelerate autoregressive text-to-image synthesis.<n> SJD-PAC employs a proactive drafting strategy to improve local acceptance rates in challenging high-entropy regions.<n>Experiments on standard text-to-image benchmarks demonstrate that SJD-PAC achieves a $3.8times$ speedup with lossless image quality.
• Score: 13.295759874474767
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Speculative Jacobi Decoding (SJD) offers a draft-model-free approach to accelerate autoregressive text-to-image synthesis. However, the high-entropy nature of visual generation yields low draft-token acceptance rates in complex regions, creating a bottleneck that severely limits overall throughput. To overcome this, we introduce SJD-PAC, an enhanced SJD framework. First, SJD-PAC employs a proactive drafting strategy to improve local acceptance rates in these challenging high-entropy regions. Second, we introduce an adaptive continuation mechanism that sustains sequence validation after an initial rejection, bypassing the need for full resampling. Working in tandem, these optimizations significantly increase the average acceptance length per step, boosting inference speed while strictly preserving the target distribution. Experiments on standard text-to-image benchmarks demonstrate that SJD-PAC achieves a $3.8\times$ speedup with lossless image quality.

Related papers

• ResiSpec: Enhancing Multi-Candidate Speculative Sampling via Residual Distribution Shaping [79.1905186547363]
  ResiSpec is a framework that strategically reforms the proposal distribution during verification.<n>It prevents candidate obsolescence and achieves up to 1.92$times$ speedup over state-of-the-art multi-candidate methods.
  arXiv  Detail & Related papers  (2026-08-25T11:25:55Z)
• CSD: Content-aware Speculative Decoding for Efficient Image Generation [42.65888524021955]
  Speculative decoding (SD) has emerged as a key solution to accelerate the inference of autoregressive models.<n>We propose a novel content-aware speculative decoding algorithm, termed CSD, which integrates an entropy-based probability relaxation mechanism.<n> Experiments conducted on Lumina-mGPT and Janus-Pro demonstrate that the superiority of the proposed CSD.
  arXiv  Detail & Related papers  (2026-06-26T08:12:42Z)
• Where to Refine, When to Stop: Rethinking Redundancy via Latent Discrepancy for Efficient Visual Autoregressive Generation [50.139984798361375]
  We propose a training-free framework that removes redundancy via latent discrepancy.<n> LD-Pruning substantially reduces inference while maintaining high generation quality, achieving up to 2.35x speedup on Infinity-8B.
  arXiv  Detail & Related papers  (2026-05-29T19:34:39Z)
• SJD-VP: Speculative Jacobi Decoding with Verification Prediction for Autoregressive Image Generation [64.43642479951576]
  Speculative Jacobi Decoding (SJD) has emerged as a promising method for accelerating autoregressive image generation.<n>We propose a novel Speculative Jacobi Decoding with Verification Prediction (SJD-VP)<n>Our SJD-VP is plug-and-play and can be seamlessly integrated into existing SJD methods.
  arXiv  Detail & Related papers  (2026-03-28T03:59:03Z)
• PACER: Blockwise Pre-verification for Speculative Decoding with Adaptive Length [21.738896310075678]
  Speculative decoding (SD) is a powerful technique for accelerating the inference process of large language models (LLMs)<n>We propose Pacer, a novel approach that dynamically controls draft length using a lightweight, trainable pre-verification layer.<n>Our results demonstrate that Pacer achieves up to 2.66x Speedup over autoregressive decoding and consistently outperforms standard speculative decoding.
  arXiv  Detail & Related papers  (2026-02-01T15:12:38Z)
• Multi-Scale Local Speculative Decoding for Image Generation [10.239314110594249]
  We introduce Multi-Scale Local Speculative Decoding (MuLo-SD)<n>MuLo-SD combines multi-resolution drafting with spatially informed verification to accelerate AR image generation.<n>We demonstrate that MuLo-SD achieves substantial speedups up to $mathbf1.7times$.
  arXiv  Detail & Related papers  (2026-01-08T17:39:35Z)
• SJD++: Improved Speculative Jacobi Decoding for Training-free Acceleration of Discrete Auto-regressive Text-to-Image Generation [92.34355601866629]
  Large autoregressive models can generate high-quality, high-resolution images but suffer from slow generation speed.<n>We propose Speculative Jacobi Decoding++ (SJD++), a training-free probabilistic parallel decoding algorithm.<n>SJD++ performs multi-token prediction in each forward pass, drastically reducing generation steps.
  arXiv  Detail & Related papers  (2025-12-08T12:36:43Z)
• MC-SJD : Maximal Coupling Speculative Jacobi Decoding for Autoregressive Visual Generation Acceleration [14.582793306013615]
  We propose MC-SJD, a training-free parallel decoding framework designed to accelerate AR visual generation.<n>We show that MC-SJD substantially accelerates standard Speculative Jacobi Decoding (SJD) by maximizing the probability of sampling identical draft tokens.<n>Remarkably, this method requires only a single-line modification to the existing algorithm, yet substantial performance gains.
  arXiv  Detail & Related papers  (2025-10-28T09:26:27Z)
• Rebalancing Contrastive Alignment with Bottlenecked Semantic Increments in Text-Video Retrieval [48.85977777168096]
  Gap-Aware Retrieval framework introduces a learnable, pair-specific increment $Delta_ij$ between text $t_i$ and video $v_j$.<n>A lightweight neural module conditioned on the semantic gap couples increments across batches for structure-aware correction.<n>Experiments on four benchmarks demonstrate that GARE consistently improves alignment accuracy and robustness.
  arXiv  Detail & Related papers  (2025-05-18T17:18:06Z)
• Accelerating Auto-regressive Text-to-Image Generation with Training-free Speculative Jacobi Decoding [60.188309982690335]
  We propose a training-free probabilistic parallel decoding algorithm, Speculative Jacobi Decoding (SJD)<n>SJD accelerates the inference of auto-regressive text-to-image generation while maintaining the randomness in sampling-based token decoding.<n>Specifically, SJD facilitates the model to predict multiple tokens at each step and accepts tokens based on the probabilistic criterion.
  arXiv  Detail & Related papers  (2024-10-02T16:05:27Z)
• Cascade Reward Sampling for Efficient Decoding-Time Alignment [17.278488115500615]
  We introduce Cascade Reward Sampling (CARDS) to resolve both efficiency in decoding-time alignment.<n>CARDS minimizes redundant computations of both large language models (LLMs) and reward models (RMs)
  arXiv  Detail & Related papers  (2024-06-24T04:08:35Z)
• Semi-Autoregressive Image Captioning [153.9658053662605]
  Current state-of-the-art approaches for image captioning typically adopt an autoregressive manner. Non-autoregressive image captioning with continuous iterative refinement can achieve comparable performance to the autoregressive counterparts with a considerable acceleration. We propose a novel two-stage framework, referred to as Semi-Autoregressive Image Captioning (SAIC) to make a better trade-off between performance and speed.
  arXiv  Detail & Related papers  (2021-10-11T15:11:54Z)

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.