Adapting Diffusion Language Models for Lossless Pixel-Level Image Transmission

2026-06-04

AI Summary

This work addresses the challenge of achieving pixel-level lossless image transmission over noisy channels by jointly optimizing source modeling fidelity and channel reliability. The authors propose DDM-SSCC, a discrete diffusion model-based separated source-channel coding framework that recovers pixel tokens using a diffusion language model and enables parallel multi-token decoding with progressive reconstruction through synchronized backward arithmetic coding under a bidirectional attention mechanism. Key innovations include Halton sequence-guided denoising order for enhanced spatial coverage, mask ratio-aware cosine noise scheduling for adaptive denoising dynamics, and a lightweight temperature calibration module to improve probability estimation accuracy. Experiments demonstrate that DDM-SSCC significantly outperforms existing lossless and semantic communication methods in exact recovery performance across CIFAR10, DIV2K-LR-X4, and Kodak datasets under both AWGN and Rayleigh fading channels.

Abstract

Lossless pixel-level image transmission is a fundamental regime beyond semantic communications, because exact recovery requires both accurate symbol probability modeling and reliable delivery over noisy channels. This paper proposes DDM-SSCC, a discrete-diffusion-model-based separate source-channel coding framework for lossless image transmission. Different from raster-order autoregressive coding, the proposed source codec adapts a diffusion language model to pixel-token restoration and performs synchronized reverse arithmetic coding under bidirectional attention, allowing multiple masked tokens to be coded within one reverse denoising step. This progressive restoration process also yields a more favorable source representation for noisy transmission, since newly restored tokens can serve as bidirectional context in subsequent denoising steps. To bridge the gap between generation-oriented masked denoising and lossless arithmetic coding, we further introduce a Halton-guided denoising order, a mask-ratio-aware cosine schedule, and a lightweight temperature calibration module. These designs respectively improve spatial coverage, adapt the denoising pace to context reliability, and calibrate the probability tables used by arithmetic coding. Experiments on CIFAR10, DIV2K-LR-X4, and Kodak over additive white Gaussian noise and Rayleigh fading channels show that DDM-SSCC achieves better exact-recovery performance than representative lossless and semantic communication baselines, while ablation studies verify the effectiveness of the proposed denoising order, schedule, and calibration modules.

Problem

• lossless image transmission
• pixel-level recovery
• noisy channels
• source-channel coding
• exact recovery

Innovation

• diffusion language model
• lossless image transmission
• separate source-channel coding
• reverse arithmetic coding
• masked denoising