paper-with-me

Papers

Time-Annealed Perturbation Sampling: Diverse Generation for Diffusion Language Models

2026-01-30 · Jingxuan Wu, Zhenglin Wan, Xingrui Yu, Yuzhe Yang, Yiqiao Huang, Ivor Tsang, Yang You arxiv

Diffusion language models (Diffusion-LMs) introduce an explicit temporal dimension into text generation, yet how this structure can be leveraged to control generation diversity for exploring multiple valid semantic or reasoning paths remains underexplored. In this paper, we show that Diffusion-LMs, like diffusion models in image generation, exhibit a temporal division of labor: early denoising steps largely determine the global semantic structure, while later steps focus on local lexical refinement. Building on this insight, we propose Time-Annealed Perturbation Sampling (TAPS), a training-free inference strategy that encourages semantic branching early in the diffusion process while progressively reducing perturbations to preserve fluency and instruction adherence. TAPS is compatible with both non-autoregressive and semi-autoregressive Diffusion backbones, demonstrated on LLaDA and TraDo in our paper, and consistently improves output diversity across creative writing and reasoning benchmarks without compromising generation quality.

📄 PDF Abstract BibTeX arXiv:2601.22629

Code (0)

등록된 구현이 없습니다.

Tasks

Image GenerationText Generation

Similar Papers 제목 키워드 기반

Annealed Relaxation of Speculative Decoding for Faster Autoregressive Image Generation

2026-01-14 · Xingyao Li, Fengzhuo Zhang, Cunxiao Du, Hui Ji arxiv

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 atte…

Image Generation

Flow Annealed Importance Sampling Bootstrap meets Differentiable Particle Physics

2024-11-25 · Annalena Kofler, Vincent Stimper, Mikhail Mikhasenko, Michael Kagan 외

High-energy physics requires the generation of large numbers of simulated data samples from complex but analytically tractable distributions called matrix elements. Surrogate models, such as normalizing flows, are gainin…

Computational Efficiency

CADS: Unleashing the Diversity of Diffusion Models through Condition-Annealed Sampling

2023-10-26 · Seyedmorteza Sadat, Jakob Buhmann, Derek Bradley, Otmar Hilliges 외

While conditional diffusion models are known to have good coverage of the data distribution, they still face limitations in output diversity, particularly when sampled with a high classifier-free guidance scale for optim…

AttributeDiversityImage Generation

Adversarial score matching and improved sampling for image generation

2020-09-11 · ICLR 2021 1 · Alexia Jolicoeur-Martineau, Rémi Piché-Taillefer, Rémi Tachet des Combes, Ioannis Mitliagkas

Denoising Score Matching with Annealed Langevin Sampling (DSM-ALS) has recently found success in generative modeling. The approach works by first training a neural network to estimate the score of a distribution, and the…

DenoisingImage Generation

Product of Experts for Visual Generation

2025-06-10 · Yunzhi Zhang, Carson Murtuza-Lanier, Zizhang Li, Yilun Du 외

Modern neural models capture rich priors and have complementary knowledge over shared data domains, e.g., images and videos. Integrating diverse knowledge from multiple sources -- including visual generative models, visu…