{"paper":{"title":"VT-Bench: A Unified Benchmark for Visual-Tabular Multi-Modal Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"VT-Bench is the first unified benchmark to standardize evaluation of models that combine images with tabular data.","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Kun-Yang Yu, Lan-Zhe Guo, Xin-Yue Zhang, Yu-Feng Li, Zhi Zhou, Zi-Jian Cheng, Zi-Yi Jia","submitted_at":"2026-05-03T09:33:08Z","abstract_excerpt":"Multi-model learning has attracted great attention in visual-text tasks. However, visual-tabular data, which plays a pivotal role in high-stakes domains like healthcare and industry, remains underexplored. In this paper, we introduce \\textit{VT-Bench}, the first unified benchmark for standardizing vision-tabular discriminative prediction and generative reasoning tasks. VT-Bench aggregates 14 datasets across 9 domains (medical-centric, while covering pets, media, and transportation) with over 756K samples. We evaluate 23 representative models, including unimodal experts, specialized visual-tabu"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"VT-Bench is the first unified benchmark for standardizing vision-tabular discriminative prediction and generative reasoning tasks, aggregating 14 datasets across 9 domains with over 756K samples and evaluating 23 models to highlight substantial challenges of visual-tabular learning.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The 14 chosen datasets and the evaluation setup for 23 models accurately represent the core difficulties of visual-tabular multi-modal learning without selection bias or incomplete coverage of real-world use cases.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"VT-Bench is the first unified benchmark aggregating 14 visual-tabular datasets with over 756K samples and evaluating 23 models to expose challenges in this multi-modal area.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"VT-Bench is the first unified benchmark to standardize evaluation of models that combine images with tabular data.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"630166a41d53b8720d1a3727ee152abc8e7be4fda30354d9950fdbc7dcf2848f"},"source":{"id":"2605.08146","kind":"arxiv","version":2},"verdict":{"id":"c3d06200-9728-4f77-bb15-ff51847de43d","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-12T01:27:29.175402Z","strongest_claim":"VT-Bench is the first unified benchmark for standardizing vision-tabular discriminative prediction and generative reasoning tasks, aggregating 14 datasets across 9 domains with over 756K samples and evaluating 23 models to highlight substantial challenges of visual-tabular learning.","one_line_summary":"VT-Bench is the first unified benchmark aggregating 14 visual-tabular datasets with over 756K samples and evaluating 23 models to expose challenges in this multi-modal area.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The 14 chosen datasets and the evaluation setup for 23 models accurately represent the core difficulties of visual-tabular multi-modal learning without selection bias or incomplete coverage of real-world use cases.","pith_extraction_headline":"VT-Bench is the first unified benchmark to standardize evaluation of models that combine images with tabular data."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.08146/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_compliance","ran_at":"2026-05-19T16:58:08.700136Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"40632a24a8bca50f020569bc4f8352a4cefa842bf25027e0500b91050bdce321"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}