Think While You Generate: Discrete Diffusion with Planned Denoising

2024-10-08

arXiv.org

AI Summary

To address the low denoising efficiency and reconstruction quality degradation caused by sequential randomness in discrete diffusion models, this paper proposes a “planning-based denoising” framework that decouples the process into a learnable position planner and a local denoiser, enabling on-demand identification of critical positions and precise token/image restoration. This two-stage paradigm overcomes inherent limitations of conventional mask diffusion—namely, uniform or random denoising—by introducing iterative adaptive mask selection and joint token/image training. Evaluated on text8, OpenWebText, and ImageNet 256×256, our method significantly outperforms state-of-the-art mask diffusion models: it achieves language modeling perplexity closely approaching autoregressive baselines while simultaneously improving both inference efficiency and fidelity in image generation.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Discrete diffusion has achieved state-of-the-art performance, outperforming or approaching autoregressive models on standard benchmarks. In this work, we introduce Discrete Diffusion with Planned Denoising (DDPD), a novel framework that separates the generation process into two models: a planner and a denoiser. At inference time, the planner selects which positions to denoise next by identifying the most corrupted positions in need of denoising, including both initially corrupted and those requiring additional refinement. This plan-and-denoise approach enables more efficient reconstruction during generation by iteratively identifying and denoising corruptions in the optimal order. DDPD outperforms traditional denoiser-only mask diffusion methods, achieving superior results on language modeling benchmarks such as text8, OpenWebText, and token-based image generation on ImageNet $256 imes 256$. Notably, in language modeling, DDPD significantly reduces the performance gap between diffusion-based and autoregressive methods in terms of generative perplexity. Code is available at https://github.com/liusulin/DDPD.

Problem

• Improving discrete diffusion models with planned denoising
• Reducing performance gap between diffusion and autoregressive methods
• Enhancing efficiency in language and image generation tasks

Innovation

• Planner and denoiser separate framework
• Optimal order denoising for efficiency
• Outperforms traditional mask diffusion methods