paper-with-me

Papers

Think While You Generate: Discrete Diffusion with Planned Denoising

2024-10-08 · Sulin Liu, Juno Nam, Andrew Campbell, Hannes Stärk, Yilun Xu, Tommi Jaakkola, Rafael Gómez-Bombarelli

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 $256 \times 256$. 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.

📄 PDF Abstract BibTeX arXiv:2410.06264

Code (1)

liusulin/ddpd 공식 구현 pytorch

Tasks

DenoisingImage GenerationLanguage ModelingLanguage Modelling

Methods 이 논문이 사용한 방법론

Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

Planned Diffusion

2025-10-20 · Daniel Israel, Tian Jin, Ellie Cheng, Guy Van den Broeck 외 arxiv

Most large language models are autoregressive: they generate tokens one at a time. Discrete diffusion language models can generate multiple tokens in parallel, but sampling from them requires a denoising order: a strateg…

Instruction Following

Planner Aware Path Learning in Diffusion Language Models Training

2025-09-27 · Fred Zhangzhi Peng, Zachary Bezemek, Jarrid Rector-Brooks, Shuibai Zhang 외 arxiv

Diffusion language models have emerged as a powerful alternative to autoregressive models, enabling fast inference through more flexible and parallel generation paths. This flexibility of sampling is unlocked by new engi…

Code GenerationText Generation

DiffuSpeech: Silent Thought, Spoken Answer via Unified Speech-Text Diffusion

2026-01-30 · Yuxuan Lou, Ziming Wu, Yaochen Wang, Yong Liu 외 arxiv

Current speech language models generate responses directly without explicit reasoning, leading to errors that cannot be corrected once audio is produced. We introduce \textbf{``Silent Thought, Spoken Answer''} -- a parad…

Stop-Think-AutoRegress: Language Modeling with Latent Diffusion Planning

2026-02-24 · Justin Lovelace, Christian Belardi, Sofian Zalouk, Adhitya Polavaram 외 arxiv

The Stop-Think-AutoRegress Language Diffusion Model (STAR-LDM) integrates latent diffusion planning with autoregressive generation. Unlike conventional autoregressive language models limited to token-by-token decisions, …

CoDAR: Continuous Diffusion Language Models are More Powerful Than You Think

2026-03-03 · Junzhe Shen, Jieru Zhao, Ziwei He, Zhouhan Lin arxiv

We study why continuous diffusion language models (DLMs) have lagged behind discrete diffusion approaches despite their appealing continuous generative dynamics. Under a controlled token--recovery study, we identify toke…