{"paper":{"title":"Dual-Diffusional Generative Fashion Recommendation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A dual-diffusion Transformer generates both fashion item images and textual descriptions for personalized recommendations.","cross_cats":["cs.MM"],"primary_cat":"cs.IR","authors_text":"Lei Wu, Mingzhe Yu, Qianru Sun, Yunshan Ma","submitted_at":"2026-05-17T09:52:18Z","abstract_excerpt":"Personalized generative recommender systems have emerged as a promising solution for fashion recommendation. However, existing methods primarily rely on implicit visual embeddings from historical interactions, which often contain preference-irrelevant information and result in insufficient user behavior modeling. Moreover, these models typically generate only item images, providing limited interpretability. To address these limitations, we propose DualFashion, a Dual-Diffusional Generative Fashion Recommendation Architecture that jointly models image and text modalities for personalized and ex"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"DualFashion achieves strong performance in behavior modeling, interpretability, and efficiency compared to state-of-the-art methods on iFashion and Polyvore-U across Personalized Fill-in-the-Blank and Generative Outfit Recommendation tasks.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That conditioning the dual-diffusion Transformer on structured attribute-level captions and visual outfit information from historical interactions sufficiently removes preference-irrelevant information and accurately models user behavior.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"DualFashion introduces a dual-diffusion Transformer with image and text branches that generates both visual items and semantic descriptions for explainable personalized fashion recommendation.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A dual-diffusion Transformer generates both fashion item images and textual descriptions for personalized recommendations.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"a8c7ffbe74c348f86522423d67401ea2d7bdeee30569760e1fbbf9a960dcfdf2"},"source":{"id":"2605.17357","kind":"arxiv","version":1},"verdict":{"id":"26e75da6-216c-4282-8090-d3b4cf49502c","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T23:09:21.779354Z","strongest_claim":"DualFashion achieves strong performance in behavior modeling, interpretability, and efficiency compared to state-of-the-art methods on iFashion and Polyvore-U across Personalized Fill-in-the-Blank and Generative Outfit Recommendation tasks.","one_line_summary":"DualFashion introduces a dual-diffusion Transformer with image and text branches that generates both visual items and semantic descriptions for explainable personalized fashion recommendation.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That conditioning the dual-diffusion Transformer on structured attribute-level captions and visual outfit information from historical interactions sufficiently removes preference-irrelevant information and accurately models user behavior.","pith_extraction_headline":"A dual-diffusion Transformer generates both fashion item images and textual descriptions for personalized recommendations."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.17357/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T23:31:20.080331Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T23:21:01.446602Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T21:41:57.789097Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T21:33:23.722598Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"57848955afa04c08136c165c9a3d842f1530ca06982e589fa69050b81eef557a"},"references":{"count":52,"sample":[{"doi":"","year":2019,"title":"Andrew Brock, Jeff Donahue, and Karen Simonyan. 2019. Large Scale GAN Training for High Fidelity Natural Image Synthesis. InICLR. OpenReview.net","work_id":"28d92ff2-5a24-48e9-a1c2-cecda8e69225","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Miaomiao Cai, Lei Chen, Yifan Wang, Haoyue Bai, Peijie Sun, Le Wu, Min Zhang, and Meng Wang. 2024. Popularity-Aware Alignment and Contrast for Mitigating Popularity Bias. InKDD. ACM, 187–198","work_id":"863cf328-bb6d-4c78-ad3d-a13e37737248","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2026,"title":"Miaomiao Cai, Zhijie Zhang, Junfeng Fang, Zhiyong Cheng, Xiang Wang, and Meng Wang. 2026. RMBRec: Robust Multi-Behavior Recommendation towards Target Behaviors. InWWW. ACM, 6731–6742","work_id":"5360b533-848e-4f5b-b034-4fff62bbf9af","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2019,"title":"Wen Chen, Pipei Huang, Jiaming Xu, Xin Guo, Cheng Guo, Fei Sun, Chao Li, Andreas Pfadler, Huan Zhao, and Binqiang Zhao. 2019. POG: Personalized Outfit Generation for Fashion Recommendation at Alibaba ","work_id":"833050b6-bb0b-4287-a43a-23ee9f0d3239","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Hao Cheng, Shuo Wang, Wensheng Lu, Wei Zhang, Mingyang Zhou, Kezhong Lu, and Hao Liao. 2023. Explainable Recommendation with Personalized Re- view Retrieval and Aspect Learning. InACL (1). Association","work_id":"30b60157-a2e0-4e56-ae17-663a50589a7d","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":52,"snapshot_sha256":"2f66d9dc46703c3e3b189a4ba9bed3082ec49c3eebb38aed5e5b20639cdcddde","internal_anchors":5},"formal_canon":{"evidence_count":1,"snapshot_sha256":"fbb149433111e7067a89ffe8484d3cdf9f59eecc735eb00b191113da02b0c83b"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}