{"ID":2865878,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.22728","arxiv_id":"2509.22728","title":"Prompt-aware classifier free guidance for diffusion models","abstract":"Diffusion models have achieved remarkable progress in image and audio generation, largely due to Classifier-Free Guidance. However, the choice of guidance scale remains underexplored: a fixed scale often fails to generalize across prompts of varying complexity, leading to oversaturation or weak alignment. We address this gap by introducing a prompt-aware framework that predicts scale-dependent quality and selects the optimal guidance at inference. Specifically, we construct a large synthetic dataset by generating samples under multiple scales and scoring them with reliable evaluation metrics. A lightweight predictor, conditioned on semantic embeddings and linguistic complexity, estimates multi-metric quality curves and determines the best scale via a utility function with regularization. Experiments on MSCOCO~2014 and AudioCaps show consistent improvements over vanilla CFG, enhancing fidelity, alignment, and perceptual preference. This work demonstrates that prompt-aware scale selection provides an effective, training-free enhancement for pretrained diffusion backbones.","short_abstract":"Diffusion models have achieved remarkable progress in image and audio generation, largely due to Classifier-Free Guidance. However, the choice of guidance scale remains underexplored: a fixed scale often fails to generalize across prompts of varying complexity, leading to oversaturation or weak alignment. We address th...","url_abs":"https://arxiv.org/abs/2509.22728","url_pdf":"https://arxiv.org/pdf/2509.22728v2","authors":"[\"Xuanhao Zhang\",\"Chang Li\"]","published":"2025-09-25T09:16:25Z","proceeding":"cs.SD","tasks":"[\"cs.SD\",\"cs.AI\",\"cs.MM\",\"eess.AS\"]","methods":"[\"Diffusion Model\"]","has_code":false}