Related papers: Let It Go or Learn to Self-Correct: Continuous Diffusion for Constrained Discrete Tasks

Let It Go or Learn to Self-Correct: Continuous Diffusion for Constrained Discrete Tasks

URL: http://arxiv.org/abs/2609.09009v1
Date: Tue, 08 Sep 2026 16:45:52 GMT
Title: Let It Go or Learn to Self-Correct: Continuous Diffusion for Constrained Discrete Tasks
Abstract summary: Denoising Diffusion Probabilistic Models generate samples by repeatedly denoising while keeping each update close to the current noisy state.<n>We show that standard diffusion sampling may preserve early mistakes, even when the model's clean predictions are informative.<n>We introduce self-correction training, which exposes the model to its own predictions, improving robustness to errors that arise during inference.
Score: 16.76819088768813
License: http://creativecommons.org/licenses/by/4.0/
Abstract: Denoising Diffusion Probabilistic Models (DDPMs) generate samples by starting from noise and repeatedly denoising while keeping each update close to the current noisy state. This behavior is effective in many continuous domains, but its role is less clear for globally constrained discrete tasks, such as Sudoku, graph connectivity, Latin squares, and N-queens. In such settings, early discrete errors can be difficult to undo. As a result, standard diffusion sampling may preserve early mistakes, even when the model's clean predictions are informative. We compare standard samplers to sampling directly from the model's clean prediction. Without retraining, this single change improves Sudoku validity from 31% to 95%, with consistent gains across the other discrete tasks. We hypothesize that staying close to the current noisy state is harmful because the reverse trajectory can drift off the forward noising distribution the model was trained on. To reduce this train-test mismatch, we further introduce self-correction training, which exposes the model to its own predictions, improving robustness to errors that arise during inference. This substantially improves the performance of standard samplers. Our results suggest that continuous diffusion models can learn nontrivial global constraints, but discrete reasoning tasks require better alignment between training and inference: either through samplers that reduce commitment to early decisions, or through training that teaches the model to correct its own inference-time errors.

Related papers

FastDiSS: Few-step Match Many-step Diffusion Language Model on Sequence-to-Sequence Generation--Full Version [4.099184871977765]
Self-conditioning degrades in one-step sampling for fast inference.<n>We propose a training framework thatperturbs the self-conditioning signal to match inference noise.<n>Our framework surpasses standard continuous diffusion models while providing up to 400x faster inference speed.
arXiv Detail & Related papers (2026-04-07T07:52:08Z)
Combating Noisy Labels through Fostering Self- and Neighbor-Consistency [120.4394402099635]
Label noise is pervasive in various real-world scenarios, posing challenges in supervised deep learning.<n>We propose a noise-robust method named Jo-SNC (textbfJoint sample selection and model regularization based on textbfSelf- and textbfNeighbor-textbfConsistency)<n>We design a self-adaptive, data-driven thresholding scheme to adjust per-class selection thresholds.
arXiv Detail & Related papers (2026-01-19T07:55:29Z)
From Denoising to Refining: A Corrective Framework for Vision-Language Diffusion Model [72.73512218682187]
We introduce ReDiff, a refining-enhanced diffusion framework that teaches the model to identify and correct its own errors.<n>Our approach features a two-stage training process: first, we instill a foundational revision capability by training the model to revise synthetic errors; second, we implement a novel online self-correction loop.<n>This mistake-driven learning endows the model with the crucial ability to revisit and refine its already generated output, effectively breaking the error cascade.
arXiv Detail & Related papers (2025-10-22T06:58:55Z)
Informed Correctors for Discrete Diffusion Models [27.295990499157814]
We propose a predictor-corrector sampling scheme for discrete diffusion models.<n>We show that our informed corrector consistently produces superior samples with fewer errors or improved FID scores.<n>Our results underscore the potential of informed correctors for fast and high-fidelity generation using discrete diffusion.
arXiv Detail & Related papers (2024-07-30T23:29:29Z)
Selective Learning: Towards Robust Calibration with Dynamic Regularization [79.92633587914659]
Miscalibration in deep learning refers to there is a discrepancy between the predicted confidence and performance. We introduce Dynamic Regularization (DReg) which aims to learn what should be learned during training thereby circumventing the confidence adjusting trade-off.
arXiv Detail & Related papers (2024-02-13T11:25:20Z)
One More Step: A Versatile Plug-and-Play Module for Rectifying Diffusion Schedule Flaws and Enhancing Low-Frequency Controls [77.42510898755037]
One More Step (OMS) is a compact network that incorporates an additional simple yet effective step during inference. OMS elevates image fidelity and harmonizes the dichotomy between training and inference, while preserving original model parameters. Once trained, various pre-trained diffusion models with the same latent domain can share the same OMS module.
arXiv Detail & Related papers (2023-11-27T12:02:42Z)
Common Diffusion Noise Schedules and Sample Steps are Flawed [7.802281665410233]
Common diffusion noise schedules do not enforce the last timestep to have zero signal-to-noise ratio. Some implementations of diffusion samplers do not start from the last timestep. We show that the flawed design causes real problems in existing implementations.
arXiv Detail & Related papers (2023-05-15T12:21:08Z)
Consistent Diffusion Models: Mitigating Sampling Drift by Learning to be Consistent [97.64313409741614]
We propose to enforce a emphconsistency property which states that predictions of the model on its own generated data are consistent across time. We show that our novel training objective yields state-of-the-art results for conditional and unconditional generation in CIFAR-10 and baseline improvements in AFHQ and FFHQ.
arXiv Detail & Related papers (2023-02-17T18:45:04Z)
Input Perturbation Reduces Exposure Bias in Diffusion Models [41.483581603727444]
We show that a long sampling chain leads to an error accumulation phenomenon, similar to the exposure bias problem in autoregressive text generation. We propose a very simple but effective training regularization, consisting in perturbing the ground truth samples to simulate the inference time prediction errors. We empirically show that, without affecting the recall and precision, the proposed input perturbation leads to a significant improvement in the sample quality.
arXiv Detail & Related papers (2023-01-27T13:34:54Z)
Mitigating Catastrophic Forgetting in Scheduled Sampling with Elastic Weight Consolidation in Neural Machine Translation [15.581515781839656]
Autoregressive models trained with maximum likelihood estimation suffer from exposure bias. We propose using Elastic Weight Consolidation as trade-off between mitigating exposure bias and retaining output quality. Experiments on two IWSLT'14 translation tasks demonstrate that our approach alleviates catastrophic forgetting and significantly improves BLEU.
arXiv Detail & Related papers (2021-09-13T20:37:58Z)
Self-Damaging Contrastive Learning [92.34124578823977]
Unlabeled data in reality is commonly imbalanced and shows a long-tail distribution. This paper proposes a principled framework called Self-Damaging Contrastive Learning to automatically balance the representation learning without knowing the classes. Our experiments show that SDCLR significantly improves not only overall accuracies but also balancedness.
arXiv Detail & Related papers (2021-06-06T00:04:49Z)

This list is automatically generated from the titles and abstracts of the papers in this site.