{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:NTBX7Z4PIVBTCY5UDHGX3454T2","short_pith_number":"pith:NTBX7Z4P","schema_version":"1.0","canonical_sha256":"6cc37fe78f45433163b419cd7df3bc9e8f3d8c245b5fe39684b2164be2e0f80a","source":{"kind":"arxiv","id":"2504.15279","version":1},"attestation_state":"computed","paper":{"title":"VisuLogic: A Benchmark for Evaluating Visual Reasoning in Multi-modal Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aijun Yang, Houqiang Li, Jiahao Wang, Jifeng Dai, Jinguo Zhu, Lewei Lu, Weiye Xu, Weiyun Wang, Wengang Zhou, Wenhai Wang, Xiaohua Wang, Xizhou Zhu, Zhe Chen","submitted_at":"2025-04-21T17:59:53Z","abstract_excerpt":"Visual reasoning is a core component of human intelligence and a critical capability for advanced multimodal models. Yet current reasoning evaluations of multimodal large language models (MLLMs) often rely on text descriptions and allow language-based reasoning shortcuts, failing to measure genuine vision-centric reasoning. To address this, we introduce VisuLogic: a benchmark of 1,000 human-verified problems across six categories (e.g., quantitative shifts, spatial relations, attribute comparisons). These various types of questions can be evaluated to assess the visual reasoning capabilities o"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2504.15279","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-04-21T17:59:53Z","cross_cats_sorted":[],"title_canon_sha256":"8a06505234690395cf5a8c91b38f895b4d00a6a3f49397a42b4b78f33d9d6445","abstract_canon_sha256":"61c164acb1140c3b80072cc74a63362f95ec171ceb21ed26617c626bcbca2913"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:51:59.226504Z","signature_b64":"1bR4AMeeIU2k77W+pTy9tRxDyibo6fK1M2wjT+2e/8Hma2MQs+iVGNN7lXBk/bMYhHNUV3GCTDris3s+F9Z2Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6cc37fe78f45433163b419cd7df3bc9e8f3d8c245b5fe39684b2164be2e0f80a","last_reissued_at":"2026-07-05T10:51:59.226003Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:51:59.226003Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VisuLogic: A Benchmark for Evaluating Visual Reasoning in Multi-modal Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aijun Yang, Houqiang Li, Jiahao Wang, Jifeng Dai, Jinguo Zhu, Lewei Lu, Weiye Xu, Weiyun Wang, Wengang Zhou, Wenhai Wang, Xiaohua Wang, Xizhou Zhu, Zhe Chen","submitted_at":"2025-04-21T17:59:53Z","abstract_excerpt":"Visual reasoning is a core component of human intelligence and a critical capability for advanced multimodal models. Yet current reasoning evaluations of multimodal large language models (MLLMs) often rely on text descriptions and allow language-based reasoning shortcuts, failing to measure genuine vision-centric reasoning. To address this, we introduce VisuLogic: a benchmark of 1,000 human-verified problems across six categories (e.g., quantitative shifts, spatial relations, attribute comparisons). These various types of questions can be evaluated to assess the visual reasoning capabilities o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.15279","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2504.15279/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2504.15279","created_at":"2026-07-05T10:51:59.226071+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.15279v1","created_at":"2026-07-05T10:51:59.226071+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.15279","created_at":"2026-07-05T10:51:59.226071+00:00"},{"alias_kind":"pith_short_12","alias_value":"NTBX7Z4PIVBT","created_at":"2026-07-05T10:51:59.226071+00:00"},{"alias_kind":"pith_short_16","alias_value":"NTBX7Z4PIVBTCY5U","created_at":"2026-07-05T10:51:59.226071+00:00"},{"alias_kind":"pith_short_8","alias_value":"NTBX7Z4P","created_at":"2026-07-05T10:51:59.226071+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":30,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05798","citing_title":"Segmentation before Answering: Pixel Grounding for MLLM Visual Reasoning","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2606.26794","citing_title":"ReasonCLIP-58M: Visually Grounded Commonsense Reasoning Supervision for CLIP","ref_index":98,"is_internal_anchor":false},{"citing_arxiv_id":"2607.01982","citing_title":"MolSight: A Graph-Aware Vision-Language Model for Unified Chemical Image Understanding","ref_index":85,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10651","citing_title":"Kwai Keye-VL-2.0 Technical Report","ref_index":96,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08035","citing_title":"DyCo-RL: Dynamic Cross-Modal Coordination for Visual Reasoning","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00248","citing_title":"Seed2.0 Model Card: Towards Intelligence Frontier for Real-World Complexity","ref_index":123,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07639","citing_title":"MOSS-Video-Preview: Toward Real-Time Video Understanding via Cross-Attention","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09883","citing_title":"The Cartesian Shortcut: Re-evaluate Vision Reasoning in Polar Coordinate Space","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00148","citing_title":"StemBind: When MLLMs Get Lost Between Rules and Instances in Abstract Visual Reasoning","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00562","citing_title":"DeepLatent: Think with Images via Parallel Latent Visual Reasoning","ref_index":107,"is_internal_anchor":false},{"citing_arxiv_id":"2602.18600","citing_title":"MapTab: A Diagnostic Benchmark for Long-Horizon Multi-Criteria Multimodal Reasoning on Heterogeneous Topological Graphs","ref_index":87,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24763","citing_title":"Tuna-2: Pixel Embeddings Beat Vision Encoders for Multimodal Understanding and Generation","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18445","citing_title":"What's Holding Back Latent Visual Reasoning?","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18641","citing_title":"Leveraging Latent Visual Reasoning in Silence","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19461","citing_title":"Beyond Mode Collapse: Distribution Matching for Diverse Reasoning","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2511.20814","citing_title":"SPHINX: A Synthetic Environment for Visual Perception and Reasoning","ref_index":62,"is_internal_anchor":false},{"citing_arxiv_id":"2601.13606","citing_title":"ChartVerse: Scaling Chart Reasoning via Reliable Programmatic Synthesis from Scratch","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2602.07026","citing_title":"Modality Gap-Driven Subspace Alignment Training Paradigm For Multimodal Large Language Models","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2602.18600","citing_title":"MapTab: A Diagnostic Benchmark for Long-Horizon Multi-Criteria Multimodal Reasoning on Heterogeneous Topological Graphs","ref_index":87,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03893","citing_title":"FeynmanBench: Benchmarking Multimodal LLMs on Diagrammatic Physics Reasoning","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05045","citing_title":"When Relations Break: Analyzing Relation Hallucination in Vision-Language Model Under Rotation and Noise","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09883","citing_title":"The Cartesian Shortcut: Re-evaluate Vision Reasoning in Polar Coordinate Space","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24763","citing_title":"Tuna-2: Pixel Embeddings Beat Vision Encoders for Multimodal Understanding and Generation","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05045","citing_title":"When Relations Break: Analyzing Relation Hallucination in Vision-Language Model Under Rotation and Noise","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20012","citing_title":"EmbodiedMidtrain: Bridging the Gap between Vision-Language Models and Vision-Language-Action Models via Mid-training","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NTBX7Z4PIVBTCY5UDHGX3454T2","json":"https://pith.science/pith/NTBX7Z4PIVBTCY5UDHGX3454T2.json","graph_json":"https://pith.science/api/pith-number/NTBX7Z4PIVBTCY5UDHGX3454T2/graph.json","events_json":"https://pith.science/api/pith-number/NTBX7Z4PIVBTCY5UDHGX3454T2/events.json","paper":"https://pith.science/paper/NTBX7Z4P"},"agent_actions":{"view_html":"https://pith.science/pith/NTBX7Z4PIVBTCY5UDHGX3454T2","download_json":"https://pith.science/pith/NTBX7Z4PIVBTCY5UDHGX3454T2.json","view_paper":"https://pith.science/paper/NTBX7Z4P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.15279&json=true","fetch_graph":"https://pith.science/api/pith-number/NTBX7Z4PIVBTCY5UDHGX3454T2/graph.json","fetch_events":"https://pith.science/api/pith-number/NTBX7Z4PIVBTCY5UDHGX3454T2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NTBX7Z4PIVBTCY5UDHGX3454T2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NTBX7Z4PIVBTCY5UDHGX3454T2/action/storage_attestation","attest_author":"https://pith.science/pith/NTBX7Z4PIVBTCY5UDHGX3454T2/action/author_attestation","sign_citation":"https://pith.science/pith/NTBX7Z4PIVBTCY5UDHGX3454T2/action/citation_signature","submit_replication":"https://pith.science/pith/NTBX7Z4PIVBTCY5UDHGX3454T2/action/replication_record"}},"created_at":"2026-07-05T10:51:59.226071+00:00","updated_at":"2026-07-05T10:51:59.226071+00:00"}