Target Concrete Score Matching: A Holistic Framework for Discrete Diffusion

• URL: http://arxiv.org/abs/2504.16431v1
• Date: Wed, 23 Apr 2025 05:32:58 GMT
• Title: Target Concrete Score Matching: A Holistic Framework for Discrete Diffusion
• Authors: Ruixiang Zhang, Shuangfei Zhai, Yizhe Zhang, James Thornton, Zijing Ou, Joshua Susskind, Navdeep Jaitly,
• Abstract summary: We present Target Concrete Score Matching (TCSM), a novel and versatile objective for training and fine-tuning discrete diffusion models.<n>TCSM supports pre-training discrete diffusion models directly from data samples, and many existing discrete diffusion approaches naturally emerge as special cases.<n>The same TCSM objective extends to post-training of discrete diffusion models, including fine-tuning using reward functions or preference data, and distillation of knowledge from pre-trained autoregressive models.
• Score: 20.440337013897658
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Discrete diffusion is a promising framework for modeling and generating discrete data. In this work, we present Target Concrete Score Matching (TCSM), a novel and versatile objective for training and fine-tuning discrete diffusion models. TCSM provides a general framework with broad applicability. It supports pre-training discrete diffusion models directly from data samples, and many existing discrete diffusion approaches naturally emerge as special cases of our more general TCSM framework. Furthermore, the same TCSM objective extends to post-training of discrete diffusion models, including fine-tuning using reward functions or preference data, and distillation of knowledge from pre-trained autoregressive models. These new capabilities stem from the core idea of TCSM, estimating the concrete score of the target distribution, which resides in the original (clean) data space. This allows seamless integration with reward functions and pre-trained models, which inherently only operate in the clean data space rather than the noisy intermediate spaces of diffusion processes. Our experiments on language modeling tasks demonstrate that TCSM matches or surpasses current methods. Additionally, TCSM is versatile, applicable to both pre-training and post-training scenarios, offering greater flexibility and sample efficiency.

Related papers

• Continuous Diffusion Model for Language Modeling [57.396578974401734]
  Existing continuous diffusion models for discrete data have limited performance compared to discrete approaches.<n>We propose a continuous diffusion model for language modeling that incorporates the geometry of the underlying categorical distribution.
  arXiv  Detail & Related papers  (2025-02-17T08:54:29Z)
• Unleashing the Potential of the Diffusion Model in Few-shot Semantic Segmentation [56.87049651707208]
  Few-shot Semantic has evolved into In-context tasks, morphing into a crucial element in assessing generalist segmentation models. Our initial focus lies in understanding how to facilitate interaction between the query image and the support image, resulting in the proposal of a KV fusion method within the self-attention framework. Based on our analysis, we establish a simple and effective framework named DiffewS, maximally retaining the original Latent Diffusion Model's generative framework.
  arXiv  Detail & Related papers  (2024-10-03T10:33:49Z)
• A Score-Based Density Formula, with Applications in Diffusion Generative Models [6.76974373198208]
  Score-based generative models (SGMs) have revolutionized the field of generative modeling, achieving unprecedented success in generating realistic and diverse content. Despite empirical advances, the theoretical basis for why optimizing the evidence lower bound (ELBO) on the log-likelihood is effective for training diffusion generative models, such as DDPMs, remains largely unexplored.
  arXiv  Detail & Related papers  (2024-08-29T17:59:07Z)
• Constrained Diffusion Models via Dual Training [80.03953599062365]
  Diffusion processes are prone to generating samples that reflect biases in a training dataset. We develop constrained diffusion models by imposing diffusion constraints based on desired distributions. We show that our constrained diffusion models generate new data from a mixture data distribution that achieves the optimal trade-off among objective and constraints.
  arXiv  Detail & Related papers  (2024-08-27T14:25:42Z)
• Model-Based Diffusion for Trajectory Optimization [8.943418808959494]
  We introduce Model-Based Diffusion (MBD), an optimization approach using the diffusion process to solve trajectory optimization (TO) problems without data. Although MBD does not require external data, it can be naturally integrated with data of diverse qualities to steer the diffusion process. MBD outperforms state-of-the-art reinforcement learning and sampling-based TO methods in challenging contact-rich tasks.
  arXiv  Detail & Related papers  (2024-05-28T22:14:25Z)
• Transfer Learning for Diffusion Models [43.10840361752551]
  Diffusion models consistently produce high-quality synthetic samples. They can be impractical in real-world applications due to high collection costs or associated risks. This paper introduces the Transfer Guided Diffusion Process (TGDP), a novel approach distinct from conventional finetuning and regularization methods.
  arXiv  Detail & Related papers  (2024-05-27T06:48:58Z)
• MG-TSD: Multi-Granularity Time Series Diffusion Models with Guided Learning Process [26.661721555671626]
  We introduce a novel Multi-Granularity Time Series (MG-TSD) model, which achieves state-of-the-art predictive performance. Our approach does not rely on additional external data, making it versatile and applicable across various domains.
  arXiv  Detail & Related papers  (2024-03-09T01:15:03Z)
• Guided Diffusion from Self-Supervised Diffusion Features [49.78673164423208]
  Guidance serves as a key concept in diffusion models, yet its effectiveness is often limited by the need for extra data annotation or pretraining. We propose a framework to extract guidance from, and specifically for, diffusion models.
  arXiv  Detail & Related papers  (2023-12-14T11:19:11Z)
• Variance-Preserving-Based Interpolation Diffusion Models for Speech Enhancement [53.2171981279647]
  We present a framework that encapsulates both the VP- and variance-exploding (VE)-based diffusion methods. To improve performance and ease model training, we analyze the common difficulties encountered in diffusion models. We evaluate our model against several methods using a public benchmark to showcase the effectiveness of our approach.
  arXiv  Detail & Related papers  (2023-06-14T14:22:22Z)
• Optimal Transport Model Distributional Robustness [33.24747882707421]
  Previous works have mainly focused on exploiting distributional robustness in the data space. We develop theories that enable us to learn the optimal robust center model distribution. Our framework can be seen as a probabilistic extension of Sharpness-Aware Minimization.
  arXiv  Detail & Related papers  (2023-06-07T06:15:12Z)
• Diff-Instruct: A Universal Approach for Transferring Knowledge From Pre-trained Diffusion Models [77.83923746319498]
  We propose a framework called Diff-Instruct to instruct the training of arbitrary generative models. We show that Diff-Instruct results in state-of-the-art single-step diffusion-based models. Experiments on refining GAN models show that the Diff-Instruct can consistently improve the pre-trained generators of GAN models.
  arXiv  Detail & Related papers  (2023-05-29T04:22:57Z)

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.