Annealed Relaxation of Speculative Decoding for Faster Autoregressive Image Generation

• URL: http://arxiv.org/abs/2601.09212v1
• Date: Wed, 14 Jan 2026 06:35:21 GMT
• Title: Annealed Relaxation of Speculative Decoding for Faster Autoregressive Image Generation
• Abstract summary: We propose COOL-SD, an annealed relaxation of speculative decoding built on two key insights.<n>Experiments validate the effectiveness of COOL-SD, showing consistent improvements over prior methods in speed-quality trade-offs.
• Score: 32.808855735559135
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Despite significant progress in autoregressive image generation, inference remains slow due to the sequential nature of AR models and the ambiguity of image tokens, even when using speculative decoding. Recent works attempt to address this with relaxed speculative decoding but lack theoretical grounding. In this paper, we establish the theoretical basis of relaxed SD and propose COOL-SD, an annealed relaxation of speculative decoding built on two key insights. The first analyzes the total variation (TV) distance between the target model and relaxed speculative decoding and yields an optimal resampling distribution that minimizes an upper bound of the distance. The second uses perturbation analysis to reveal an annealing behaviour in relaxed speculative decoding, motivating our annealed design. Together, these insights enable COOL-SD to generate images faster with comparable quality, or achieve better quality at similar latency. Experiments validate the effectiveness of COOL-SD, showing consistent improvements over prior methods in speed-quality trade-offs.

Related papers

• 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)
• CASCADE: Context-Aware Relaxation for Speculative Image Decoding [9.004650208708235]
  Autoregressive generation is a powerful approach for high-fidelity image synthesis.<n>Current approaches fail to achieve efficiency gains comparable to those observed in text generation.<n>We identify previously overlooked patterns in the target model's behavior that emerge naturally in tree-based speculative decoding.
  arXiv  Detail & Related papers  (2026-05-08T04:32:17Z)
• 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)
• Hawk: Leveraging Spatial Context for Faster Autoregressive Text-to-Image Generation [87.00172597953228]
  Speculative decoding has shown promise in accelerating text generation without compromising quality.<n>We introduce Hawk, a new approach that harnesses the spatial structure of images to guide the speculative model toward more accurate and efficient predictions.<n> Experimental results on multiple text-to-image benchmarks demonstrate a 1.71x speedup over standard AR models.
  arXiv  Detail & Related papers  (2025-10-29T17:43:31Z)
• Steering One-Step Diffusion Model with Fidelity-Rich Decoder for Fast Image Compression [36.10674664089876]
  SODEC is a novel single-step diffusion-based image compression model.<n>It improves fidelity resulting from over-reliance on generative priors.<n>It significantly outperforms existing methods, achieving superior rate-distortion-perception performance.
  arXiv  Detail & Related papers  (2025-08-07T02:24:03Z)
• Higher fidelity perceptual image and video compression with a latent conditioned residual denoising diffusion model [55.2480439325792]
  We propose a hybrid compression scheme optimized for perceptual quality, extending the approach of the CDC model with a decoder network.<n>We achieve up to +2dB PSNR fidelity improvements while maintaining comparable LPIPS and FID perceptual scores when compared with CDC.
  arXiv  Detail & Related papers  (2025-05-19T14:13:14Z)
• Ultra Lowrate Image Compression with Semantic Residual Coding and Compression-aware Diffusion [28.61304513668606]
  ResULIC is a residual-guided ultra lowrate image compression system.<n>It incorporates residual signals into both semantic retrieval and the diffusion-based generation process.<n>It achieves superior objective and subjective performance compared to state-of-the-art diffusion-based methods.
  arXiv  Detail & Related papers  (2025-05-13T06:51:23Z)
• CALLIC: Content Adaptive Learning for Lossless Image Compression [64.47244912937204]
  CALLIC sets a new state-of-the-art (SOTA) for learned lossless image compression.<n>We propose a content-aware autoregressive self-attention mechanism by leveraging convolutional gating operations.<n>During encoding, we decompose pre-trained layers, including depth-wise convolutions, using low-rank matrices and then adapt the incremental weights on testing image by Rate-guided Progressive Fine-Tuning (RPFT)<n>RPFT fine-tunes with gradually increasing patches that are sorted in descending order by estimated entropy, optimizing learning process and reducing adaptation time.
  arXiv  Detail & Related papers  (2024-12-23T10:41:18Z)
• Timestep-Aware Diffusion Model for Extreme Image Rescaling [47.89362819768323]
  We propose a novel framework called Timestep-Aware Diffusion Model (TADM) for extreme image rescaling.<n>TADM performs rescaling operations in the latent space of a pre-trained autoencoder.<n>It effectively leverages powerful natural image priors learned by a pre-trained text-to-image diffusion model.
  arXiv  Detail & Related papers  (2024-08-17T09:51:42Z)
• Faster Diffusion: Rethinking the Role of the Encoder for Diffusion Model Inference [95.42299246592756]
  We study the UNet encoder and empirically analyze the encoder features. We find that encoder features change minimally, whereas the decoder features exhibit substantial variations across different time-steps. We validate our approach on other tasks: text-to-video, personalized generation and reference-guided generation.
  arXiv  Detail & Related papers  (2023-12-15T08:46:43Z)
• Efficient Test-Time Adaptation for Super-Resolution with Second-Order Degradation and Reconstruction [62.955327005837475]
  Image super-resolution (SR) aims to learn a mapping from low-resolution (LR) to high-resolution (HR) using paired HR-LR training images. We present an efficient test-time adaptation framework for SR, named SRTTA, which is able to quickly adapt SR models to test domains with different/unknown degradation types.
  arXiv  Detail & Related papers  (2023-10-29T13:58:57Z)
• 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)
• Causal Contextual Prediction for Learned Image Compression [36.08393281509613]
  We propose the concept of separate entropy coding to leverage a serial decoding process for causal contextual entropy prediction in the latent space. A causal context model is proposed that separates the latents across channels and makes use of cross-channel relationships to generate highly informative contexts. We also propose a causal global prediction model, which is able to find global reference points for accurate predictions of unknown points.
  arXiv  Detail & Related papers  (2020-11-19T08:15:10Z)

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.