{"paper":{"title":"OxyEcomBench: Benchmarking Multimodal Foundation Models across E-Commerce Ecosystems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A new benchmark for e-commerce shows leading multimodal models achieve only modest performance because they lack domain-specific knowledge.","cross_cats":[],"primary_cat":"cs.DB","authors_text":"Bing Bai, Guoqing Yang, Ke Zhang, Xiaoqiang Xu, Ximan Liu, Yan Li, Yong Liu, Zhen Chen","submitted_at":"2026-05-13T08:34:10Z","abstract_excerpt":"LLMs and MLLMs have become indispensable tools across a wide range of applications. E-commerce, however, poses distinctive challenges -- including intricate domain knowledge, long-tail product evidence, heterogeneous visual data, and the interplay among multiple stakeholder roles -- that diverge substantially from the general world knowledge these models are primarily trained on, often causing a notable gap between their open-domain and e-commerce performance. To systematically quantify this gap, we introduce OxyEcomBench, a unified multimodal benchmark comprising approximately 6,300 high-qual"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Evaluations on 20 mainstream LLMs and MLLMs show that even the leading models attain modest performance and that performance gaps narrow on OxyEcomBench, suggesting that insufficient e-commerce-specific knowledge infusion mutes the advantages of advanced general-purpose models in this domain.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the 29 tasks and the four-level P0-P3 difficulty rubric, chosen with expert consensus and emphasis on visually salient cases, faithfully represent the full range of real-world e-commerce challenges without introducing selection bias.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"OxyEcomBench is a unified multimodal benchmark covering 6 capability areas and 29 tasks with authentic e-commerce data to measure how well foundation models handle real platform, merchant, and customer challenges.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A new benchmark for e-commerce shows leading multimodal models achieve only modest performance because they lack domain-specific knowledge.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"16d02d10c188b709eccd5507b22b1fffbf6a51b7df8750bb034d6fdfbb4431ec"},"source":{"id":"2605.13173","kind":"arxiv","version":1},"verdict":{"id":"3bdd9daa-01a4-47c3-bc13-6192e3bb8625","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T02:09:18.556656Z","strongest_claim":"Evaluations on 20 mainstream LLMs and MLLMs show that even the leading models attain modest performance and that performance gaps narrow on OxyEcomBench, suggesting that insufficient e-commerce-specific knowledge infusion mutes the advantages of advanced general-purpose models in this domain.","one_line_summary":"OxyEcomBench is a unified multimodal benchmark covering 6 capability areas and 29 tasks with authentic e-commerce data to measure how well foundation models handle real platform, merchant, and customer challenges.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the 29 tasks and the four-level P0-P3 difficulty rubric, chosen with expert consensus and emphasis on visually salient cases, faithfully represent the full range of real-world e-commerce challenges without introducing selection bias.","pith_extraction_headline":"A new benchmark for e-commerce shows leading multimodal models achieve only modest performance because they lack domain-specific knowledge."},"references":{"count":44,"sample":[{"doi":"","year":2023,"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","ref_index":1,"cited_arxiv_id":"2303.08774","is_internal_anchor":true},{"doi":"","year":2023,"title":"Visual instruction tuning","work_id":"9c789f5b-025e-4512-9ed2-daebb10acfc2","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond","work_id":"cbc2bb21-b6bb-46c0-80bf-107e195ffe10","ref_index":3,"cited_arxiv_id":"2308.12966","is_internal_anchor":true},{"doi":"","year":2024,"title":"Shopping MMLU: A massive multi-task online shopping benchmark for large language models","work_id":"1bfe8a42-99e6-4f4d-a8f8-70f2b7df177c","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Haoxin Wang, Xianhan Peng, Huang Cheng, Yizhe Huang, Ming Gong, Chenghan Y ang, Y ang Liu, and Jiang Lin. ECom-Bench: Can LLM agent resolve real-world e-commerce customer support issues? In Proceeding","work_id":"72aa7b29-f716-48de-96b2-42a8c3e8cadc","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":44,"snapshot_sha256":"91385b22bfb6e193a4ac4b66f942f660dcaad685b04729a415126b7489d7b6e0","internal_anchors":7},"formal_canon":{"evidence_count":1,"snapshot_sha256":"1e955c54999ec0a6612f84f26603500824970260046ce484144a9a198e3e4069"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}