{"ID":2866436,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.19876","arxiv_id":"2509.19876","title":"Adaptive User Interest Modeling via Conditioned Denoising Diffusion For Click-Through Rate Prediction","abstract":"User behavior sequences in search systems resemble \"interest fossils\", capturing genuine intent yet eroded by exposure bias, category drift, and contextual noise. Current methods predominantly follow an \"identify-aggregate\" paradigm, assuming sequences immutably reflect user preferences while overlooking the organic entanglement of noise and genuine interest. Moreover, they output static, context-agnostic representations, failing to adapt to dynamic intent shifts under varying Query-User-Item-Context conditions. To resolve this dual challenge, we propose the Contextual Diffusion Purifier (CDP). By treating category-filtered behaviors as \"contaminated observations\", CDP employs a forward noising and conditional reverse denoising process guided by cross-interaction features (Query x User x Item x Context), controllably generating pure, context-aware interest representations that dynamically evolve with scenarios. Extensive offline/online experiments demonstrate the superiority of CDP over state-of-the-art methods.","short_abstract":"User behavior sequences in search systems resemble \"interest fossils\", capturing genuine intent yet eroded by exposure bias, category drift, and contextual noise. Current methods predominantly follow an \"identify-aggregate\" paradigm, assuming sequences immutably reflect user preferences while overlooking the organic en...","url_abs":"https://arxiv.org/abs/2509.19876","url_pdf":"https://arxiv.org/pdf/2509.19876v1","authors":"[\"Qihang Zhao\",\"Xiaoyang Zheng\",\"Ben Chen\",\"Zhongbo Sun\",\"Chenyi Lei\"]","published":"2025-09-24T08:28:33Z","proceeding":"cs.IR","tasks":"[\"cs.IR\"]","methods":"[\"Diffusion Model\",\"Generative Adversarial Network\"]","has_code":false}