{"ID":2881106,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.12880","arxiv_id":"2508.12880","title":"Stochastic Self-Guidance for Training-Free Enhancement of Diffusion Models","abstract":"Classifier-free Guidance (CFG) is a widely used technique in modern diffusion models for enhancing sample quality and prompt adherence. However, through an empirical analysis on Gaussian mixture modeling with a closed-form solution, we observe a discrepancy between the suboptimal results produced by CFG and the ground truth. The model's excessive reliance on these suboptimal predictions often leads to semantic incoherence and low-quality outputs. To address this issue, we first empirically demonstrate that the model's suboptimal predictions can be effectively refined using sub-networks of the model itself. Building on this insight, we propose S$^2$-Guidance, a novel method that leverages stochastic block-dropping during the forward process to construct stochastic sub-networks, effectively guiding the model away from potential low-quality predictions and toward high-quality outputs. Extensive qualitative and quantitative experiments on text-to-image and text-to-video generation tasks demonstrate that S$^2$-Guidance delivers superior performance, consistently surpassing CFG and other advanced guidance strategies. Our code will be released.","short_abstract":"Classifier-free Guidance (CFG) is a widely used technique in modern diffusion models for enhancing sample quality and prompt adherence. However, through an empirical analysis on Gaussian mixture modeling with a closed-form solution, we observe a discrepancy between the suboptimal results produced by CFG and the ground...","url_abs":"https://arxiv.org/abs/2508.12880","url_pdf":"https://arxiv.org/pdf/2508.12880v4","authors":"[\"Chubin Chen\",\"Jiashu Zhu\",\"Xiaokun Feng\",\"Nisha Huang\",\"Chen Zhu\",\"Meiqi Wu\",\"Fangyuan Mao\",\"Jiahong Wu\",\"Xiangxiang Chu\",\"Xiu Li\"]","published":"2025-08-18T12:31:20Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Diffusion Model\"]","has_code":false}