{"ID":3083647,"CreatedAt":"2026-06-05T06:46:15.197025399Z","UpdatedAt":"2026-06-07T08:44:05.167452725Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2606.06303","arxiv_id":"2606.06303","title":"Plug-and-Play Guidance for Discrete Diffusion Models via Gradient-Informed Logit Correction","abstract":"Controllable generation with discrete diffusion models is often hindered by high computational overhead or the need for retraining. In this paper, we present \\underline{\\textbf{G}}radient-\\underline{\\textbf{I}}nformed \\underline{\\textbf{L}}ogit \\underline{\\textbf{C}}orrection (\\textbf{GILC}), a plug-and-play framework that efficiently estimates guidance signals by repurposing the pretrained denoising network as a variational proxy. To circumvent the gradient instability inherent in high-dimensional discrete spaces, we introduce a Jacobian-free mechanism that directly corrects the clean prediction logits, facilitating stable and effective guidance. Our method accommodates both differentiable and non-differentiable reward functions. Extensive experiments across DNA, protein sequence, and molecular generation tasks demonstrate that GILC achieves state-of-the-art performance without additional training, frequently outperforming fine-tuning approaches.","short_abstract":"Controllable generation with discrete diffusion models is often hindered by high computational overhead or the need for retraining. In this paper, we present \\underline{\\textbf{G}}radient-\\underline{\\textbf{I}}nformed \\underline{\\textbf{L}}ogit \\underline{\\textbf{C}}orrection (\\textbf{GILC}), a plug-and-play framework...","url_abs":"https://arxiv.org/abs/2606.06303","url_pdf":"https://arxiv.org/pdf/2606.06303v1","authors":"[\"Hongkun Dou\",\"Zike Chen\",\"Fengji Li\",\"Hongjue Li\",\"Yue Deng\"]","published":"2026-06-04T15:41:53Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\"]","methods":"[\"Diffusion Model\"]","has_code":false}