{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:W2CG4GPBFMRGXJFKFI3XQHA7A6","short_pith_number":"pith:W2CG4GPB","schema_version":"1.0","canonical_sha256":"b6846e19e12b226ba4aa2a37781c1f07b768c816cb94c5e36acd9fb5f3c62d02","source":{"kind":"arxiv","id":"2507.21924","version":1},"attestation_state":"computed","paper":{"title":"MMAT-1M: A Large Reasoning Dataset for Multimodal Agent Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Gang Zhang, Haixiao Yue, Shanshan Liu, Tianhong Gao, Weiqun Wu, Yannian Fu","submitted_at":"2025-07-29T15:39:14Z","abstract_excerpt":"Large Language Models (LLMs), enhanced through agent tuning, have demonstrated remarkable capabilities in Chain-of-Thought (CoT) and tool utilization, significantly surpassing the performance of standalone models. However, the multimodal domain still lacks a large-scale, high-quality agent tuning dataset to unlock the full potential of multimodal large language models. To bridge this gap, we introduce MMAT-1M, the first million-scale multimodal agent tuning dataset designed to support CoT, reflection, and dynamic tool usage. Our dataset is constructed through a novel four-stage data engine: 1)"},"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":"2507.21924","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-07-29T15:39:14Z","cross_cats_sorted":[],"title_canon_sha256":"6d6d134021f6e19049339f7e7ded1390fe4ec3dce2e8c7fa44b32403a25f0bc9","abstract_canon_sha256":"3546d74051ec9dbf42322bd6c3becc5d135e0756150712e39288731c2f235b26"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:45:09.904002Z","signature_b64":"oogZrS9T7y3sgsatTvlQuqATedv3ElPf31yi4ebWMAwLv7nWGu8EIMTmUxUu/qDuX3AhenRneikBIbmEOvyPDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b6846e19e12b226ba4aa2a37781c1f07b768c816cb94c5e36acd9fb5f3c62d02","last_reissued_at":"2026-07-05T11:45:09.903509Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:45:09.903509Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MMAT-1M: A Large Reasoning Dataset for Multimodal Agent Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Gang Zhang, Haixiao Yue, Shanshan Liu, Tianhong Gao, Weiqun Wu, Yannian Fu","submitted_at":"2025-07-29T15:39:14Z","abstract_excerpt":"Large Language Models (LLMs), enhanced through agent tuning, have demonstrated remarkable capabilities in Chain-of-Thought (CoT) and tool utilization, significantly surpassing the performance of standalone models. However, the multimodal domain still lacks a large-scale, high-quality agent tuning dataset to unlock the full potential of multimodal large language models. To bridge this gap, we introduce MMAT-1M, the first million-scale multimodal agent tuning dataset designed to support CoT, reflection, and dynamic tool usage. Our dataset is constructed through a novel four-stage data engine: 1)"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.21924","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/2507.21924/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":"2507.21924","created_at":"2026-07-05T11:45:09.903577+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.21924v1","created_at":"2026-07-05T11:45:09.903577+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.21924","created_at":"2026-07-05T11:45:09.903577+00:00"},{"alias_kind":"pith_short_12","alias_value":"W2CG4GPBFMRG","created_at":"2026-07-05T11:45:09.903577+00:00"},{"alias_kind":"pith_short_16","alias_value":"W2CG4GPBFMRGXJFK","created_at":"2026-07-05T11:45:09.903577+00:00"},{"alias_kind":"pith_short_8","alias_value":"W2CG4GPB","created_at":"2026-07-05T11:45:09.903577+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/W2CG4GPBFMRGXJFKFI3XQHA7A6","json":"https://pith.science/pith/W2CG4GPBFMRGXJFKFI3XQHA7A6.json","graph_json":"https://pith.science/api/pith-number/W2CG4GPBFMRGXJFKFI3XQHA7A6/graph.json","events_json":"https://pith.science/api/pith-number/W2CG4GPBFMRGXJFKFI3XQHA7A6/events.json","paper":"https://pith.science/paper/W2CG4GPB"},"agent_actions":{"view_html":"https://pith.science/pith/W2CG4GPBFMRGXJFKFI3XQHA7A6","download_json":"https://pith.science/pith/W2CG4GPBFMRGXJFKFI3XQHA7A6.json","view_paper":"https://pith.science/paper/W2CG4GPB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.21924&json=true","fetch_graph":"https://pith.science/api/pith-number/W2CG4GPBFMRGXJFKFI3XQHA7A6/graph.json","fetch_events":"https://pith.science/api/pith-number/W2CG4GPBFMRGXJFKFI3XQHA7A6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/W2CG4GPBFMRGXJFKFI3XQHA7A6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/W2CG4GPBFMRGXJFKFI3XQHA7A6/action/storage_attestation","attest_author":"https://pith.science/pith/W2CG4GPBFMRGXJFKFI3XQHA7A6/action/author_attestation","sign_citation":"https://pith.science/pith/W2CG4GPBFMRGXJFKFI3XQHA7A6/action/citation_signature","submit_replication":"https://pith.science/pith/W2CG4GPBFMRGXJFKFI3XQHA7A6/action/replication_record"}},"created_at":"2026-07-05T11:45:09.903577+00:00","updated_at":"2026-07-05T11:45:09.903577+00:00"}