{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:NATOKERNKSO376EKHORWGTZWTN","short_pith_number":"pith:NATOKERN","schema_version":"1.0","canonical_sha256":"6826e5122d549dbff88a3ba3634f369b41acafdd21f33a83d0d09e692e2bfd6e","source":{"kind":"arxiv","id":"2407.08739","version":2},"attestation_state":"computed","paper":{"title":"MAVIS: Mathematical Visual Instruction Tuning with an Automatic Data Engine","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aojun Zhou, Bin Wei, Chengzhuo Tong, Chunyuan Li, Dongzhi Jiang, Hongsheng Li, Jiaming Liu, Peng Gao, Renrui Zhang, Shanghang Zhang, Shicheng Li, Xinyu Wei, Yichi Zhang, Ziyu Guo","submitted_at":"2024-07-11T17:59:47Z","abstract_excerpt":"The mathematical capabilities of Multi-modal Large Language Models (MLLMs) remain under-explored with three areas to be improved: visual encoding of math diagrams, diagram-language alignment, and chain-of-thought (CoT) reasoning. This draws forth an urgent demand for an effective training paradigm and a large-scale, comprehensive dataset with detailed CoT rationales, which is challenging to collect and costly to annotate manually. To tackle this issue, we propose MAVIS, a MAthematical VISual instruction tuning pipeline for MLLMs, featuring an automatic data engine to efficiently create mathema"},"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":"2407.08739","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-07-11T17:59:47Z","cross_cats_sorted":[],"title_canon_sha256":"ef6e07a0965546245feb02e894730a9cbab887f62b7346a75a92e6724ed4c700","abstract_canon_sha256":"598dacda624ee94e962f48cbf8ec1ad51a6714ea21acfd07234fa61ba5fdd2f8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:29:49.658861Z","signature_b64":"rkjcPawSSs3HrvimfhJRuYQo7uAJ44AwyiAszujDj9TltKPnAXmvPUrrm5/ui/skB/U4kRFCo0q498f3c/cBCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6826e5122d549dbff88a3ba3634f369b41acafdd21f33a83d0d09e692e2bfd6e","last_reissued_at":"2026-07-05T09:29:49.658383Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:29:49.658383Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MAVIS: Mathematical Visual Instruction Tuning with an Automatic Data Engine","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aojun Zhou, Bin Wei, Chengzhuo Tong, Chunyuan Li, Dongzhi Jiang, Hongsheng Li, Jiaming Liu, Peng Gao, Renrui Zhang, Shanghang Zhang, Shicheng Li, Xinyu Wei, Yichi Zhang, Ziyu Guo","submitted_at":"2024-07-11T17:59:47Z","abstract_excerpt":"The mathematical capabilities of Multi-modal Large Language Models (MLLMs) remain under-explored with three areas to be improved: visual encoding of math diagrams, diagram-language alignment, and chain-of-thought (CoT) reasoning. This draws forth an urgent demand for an effective training paradigm and a large-scale, comprehensive dataset with detailed CoT rationales, which is challenging to collect and costly to annotate manually. To tackle this issue, we propose MAVIS, a MAthematical VISual instruction tuning pipeline for MLLMs, featuring an automatic data engine to efficiently create mathema"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.08739","kind":"arxiv","version":2},"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/2407.08739/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":"2407.08739","created_at":"2026-07-05T09:29:49.658435+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.08739v2","created_at":"2026-07-05T09:29:49.658435+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.08739","created_at":"2026-07-05T09:29:49.658435+00:00"},{"alias_kind":"pith_short_12","alias_value":"NATOKERNKSO3","created_at":"2026-07-05T09:29:49.658435+00:00"},{"alias_kind":"pith_short_16","alias_value":"NATOKERNKSO376EK","created_at":"2026-07-05T09:29:49.658435+00:00"},{"alias_kind":"pith_short_8","alias_value":"NATOKERN","created_at":"2026-07-05T09:29:49.658435+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08728","citing_title":"Artificial Intelligence for Mathematical Reasoning: An Integrated Survey of Language Models, Neuro-symbolic Systems, and Verified Discovery","ref_index":73,"is_internal_anchor":false},{"citing_arxiv_id":"2503.16549","citing_title":"MathFlow: Enhancing the Perceptual Flow of MLLMs for Visual Mathematical Problems","ref_index":80,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19852","citing_title":"Are Tools Always Beneficial? Learning to Invoke Tools Adaptively for Dual-Mode Multimodal LLM Reasoning","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2506.06856","citing_title":"Vision-EKIPL: External Knowledge-Infused Policy Learning for Visual Reasoning","ref_index":68,"is_internal_anchor":false},{"citing_arxiv_id":"2510.21122","citing_title":"NoisyGRPO: Incentivizing Multimodal CoT Reasoning via Noise Injection and Bayesian Estimation","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2403.14624","citing_title":"MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?","ref_index":70,"is_internal_anchor":false},{"citing_arxiv_id":"2512.07348","citing_title":"MICo-150K: A Comprehensive Dataset Advancing Multi-Image Composition","ref_index":98,"is_internal_anchor":false},{"citing_arxiv_id":"2411.10442","citing_title":"Enhancing the Reasoning Ability of Multimodal Large Language Models via Mixed Preference Optimization","ref_index":115,"is_internal_anchor":false},{"citing_arxiv_id":"2503.12605","citing_title":"Multimodal Chain-of-Thought Reasoning: A Comprehensive Survey","ref_index":161,"is_internal_anchor":false},{"citing_arxiv_id":"2412.10302","citing_title":"DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding","ref_index":110,"is_internal_anchor":false},{"citing_arxiv_id":"2407.07895","citing_title":"LLaVA-NeXT-Interleave: Tackling Multi-image, Video, and 3D in Large Multimodal Models","ref_index":66,"is_internal_anchor":false},{"citing_arxiv_id":"2504.10479","citing_title":"InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models","ref_index":148,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NATOKERNKSO376EKHORWGTZWTN","json":"https://pith.science/pith/NATOKERNKSO376EKHORWGTZWTN.json","graph_json":"https://pith.science/api/pith-number/NATOKERNKSO376EKHORWGTZWTN/graph.json","events_json":"https://pith.science/api/pith-number/NATOKERNKSO376EKHORWGTZWTN/events.json","paper":"https://pith.science/paper/NATOKERN"},"agent_actions":{"view_html":"https://pith.science/pith/NATOKERNKSO376EKHORWGTZWTN","download_json":"https://pith.science/pith/NATOKERNKSO376EKHORWGTZWTN.json","view_paper":"https://pith.science/paper/NATOKERN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.08739&json=true","fetch_graph":"https://pith.science/api/pith-number/NATOKERNKSO376EKHORWGTZWTN/graph.json","fetch_events":"https://pith.science/api/pith-number/NATOKERNKSO376EKHORWGTZWTN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NATOKERNKSO376EKHORWGTZWTN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NATOKERNKSO376EKHORWGTZWTN/action/storage_attestation","attest_author":"https://pith.science/pith/NATOKERNKSO376EKHORWGTZWTN/action/author_attestation","sign_citation":"https://pith.science/pith/NATOKERNKSO376EKHORWGTZWTN/action/citation_signature","submit_replication":"https://pith.science/pith/NATOKERNKSO376EKHORWGTZWTN/action/replication_record"}},"created_at":"2026-07-05T09:29:49.658435+00:00","updated_at":"2026-07-05T09:29:49.658435+00:00"}