{"ID":2852544,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.17136","arxiv_id":"2510.17136","title":"In-situ Autoguidance: Eliciting Self-Correction in Diffusion Models","abstract":"The generation of high-quality, diverse, and prompt-aligned images is a central goal in image-generating diffusion models. The popular classifier-free guidance (CFG) approach improves quality and alignment at the cost of reduced variation, creating an inherent entanglement of these effects. Recent work has successfully disentangled these properties by guiding a model with a separately trained, inferior counterpart; however, this solution introduces the considerable overhead of requiring an auxiliary model. We challenge this prerequisite by introducing In-situ Autoguidance, a method that elicits guidance from the model itself without any auxiliary components. Our approach dynamically generates an inferior prediction on the fly using a stochastic forward pass, reframing guidance as a form of inference-time self-correction. We demonstrate that this zero-cost approach is not only viable but also establishes a powerful new baseline for cost-efficient guidance, proving that the benefits of self-guidance can be achieved without external models.","short_abstract":"The generation of high-quality, diverse, and prompt-aligned images is a central goal in image-generating diffusion models. The popular classifier-free guidance (CFG) approach improves quality and alignment at the cost of reduced variation, creating an inherent entanglement of these effects. Recent work has successfully...","url_abs":"https://arxiv.org/abs/2510.17136","url_pdf":"https://arxiv.org/pdf/2510.17136v1","authors":"[\"Enhao Gu\",\"Haolin Hou\"]","published":"2025-10-20T04:06:50Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[\"Diffusion Model\"]","has_code":false}