{"ID":5935700,"CreatedAt":"2026-07-07T01:22:02.77346169Z","UpdatedAt":"2026-07-07T02:10:06.972658124Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.03397","arxiv_id":"2607.03397","title":"Efficient bias mitigation in T2I diffusion models using Concept Graphs","abstract":"Text-to-Image diffusion models often propagate harmful bias inherited from the training data. Existing bias mitigation techniques typically intervene only at the text encoder or provide inference-time guidance, often leading to generations that collapse into semantically incoherent outputs. To address these limitations, we introduce CO-ALIGN (Concept Ontology Alignment), a novel bias mitigation approach based on concept-graph alignment that operates on the model's internal concept ontology. By aligning concepts within the text encoder and denoiser, CO-ALIGN achieves substantial bias reduction while preserving generative integrity. We demonstrate the effectiveness of concept-graph alignment across three paradigms: text-encoders, denoisers and joint text-denoiser ontology alignment. CO-ALIGN outperforms the state of the art, improving fairness by $30\\%$, $ΔFID=11.4$ in image quality, $2.8\\%$ in image fidelity, all while reducing semantically incoherent outputs by $88\\%$. Beyond bias mitigation, we show that CO-ALIGN benefits other downstream tasks as well. In particular, our experiments demonstrate that better-aligned internal ontologies enhance concept unlearning robustness across multiple unlearning techniques.","short_abstract":"Text-to-Image diffusion models often propagate harmful bias inherited from the training data. Existing bias mitigation techniques typically intervene only at the text encoder or provide inference-time guidance, often leading to generations that collapse into semantically incoherent outputs. To address these limitations...","url_abs":"https://arxiv.org/abs/2607.03397","url_pdf":"https://arxiv.org/pdf/2607.03397v1","authors":"[\"Mansi\",\"Avinash Kori\",\"Francesco Leofante\"]","published":"2026-07-03T14:58:56Z","proceeding":"cs.AI","tasks":"[\"cs.AI\",\"cs.CV\"]","methods":"[\"Diffusion Model\"]","has_code":false}