{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:4LPIUM4K4ZSICKQY6IRO7PWX4Z","short_pith_number":"pith:4LPIUM4K","canonical_record":{"source":{"id":"2412.15606","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-12-20T07:00:46Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"4258df22b44040723a5bb8c47a5789ff9bc51b62bc6fbc449fb4a20c9e4ae093","abstract_canon_sha256":"6a780f055d99d88f21a3742af55e51ef83c9b5ec17d346d4109e5536c2d6c10b"},"schema_version":"1.0"},"canonical_sha256":"e2de8a338ae664812a18f222efbed7e669689ce6d9807d8e607b9d03afae0760","source":{"kind":"arxiv","id":"2412.15606","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.15606","created_at":"2026-07-05T10:08:26Z"},{"alias_kind":"arxiv_version","alias_value":"2412.15606v2","created_at":"2026-07-05T10:08:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.15606","created_at":"2026-07-05T10:08:26Z"},{"alias_kind":"pith_short_12","alias_value":"4LPIUM4K4ZSI","created_at":"2026-07-05T10:08:26Z"},{"alias_kind":"pith_short_16","alias_value":"4LPIUM4K4ZSICKQY","created_at":"2026-07-05T10:08:26Z"},{"alias_kind":"pith_short_8","alias_value":"4LPIUM4K","created_at":"2026-07-05T10:08:26Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:4LPIUM4K4ZSICKQY6IRO7PWX4Z","target":"record","payload":{"canonical_record":{"source":{"id":"2412.15606","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-12-20T07:00:46Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"4258df22b44040723a5bb8c47a5789ff9bc51b62bc6fbc449fb4a20c9e4ae093","abstract_canon_sha256":"6a780f055d99d88f21a3742af55e51ef83c9b5ec17d346d4109e5536c2d6c10b"},"schema_version":"1.0"},"canonical_sha256":"e2de8a338ae664812a18f222efbed7e669689ce6d9807d8e607b9d03afae0760","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:08:26.605843Z","signature_b64":"mptRWRE+Q643XjssjspKIsRwWc+CbSheuK4ls60mRBvq3DmVAMc5bYSZb3obot0uSXOOHZfEKgmPqyTFnImqBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e2de8a338ae664812a18f222efbed7e669689ce6d9807d8e607b9d03afae0760","last_reissued_at":"2026-07-05T10:08:26.605350Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:08:26.605350Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.15606","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:08:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BlMi1PGCNdwlU/Jzaz0GyEWR3eCGcVNW3fOrcmX+2hTE3NeFol8nW6G4R4tDQ6E18eZBARzp5NyrCYbgqHeSCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T17:55:04.862732Z"},"content_sha256":"324306679d0a9efa41fae88d6c86eb5bc721e4ab1500f7ee955ebfe23cf3b9b7","schema_version":"1.0","event_id":"sha256:324306679d0a9efa41fae88d6c86eb5bc721e4ab1500f7ee955ebfe23cf3b9b7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:4LPIUM4K4ZSICKQY6IRO7PWX4Z","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Multi-modal Agent Tuning: Building a VLM-Driven Agent for Efficient Tool Usage","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.AI","authors_text":"Bofei Zhang, Pengxiang Li, Qing Li, Song-Chun Zhu, Tao Yuan, Xiaojian Ma, Yue Fan, Yunde Jia, Yuwei Wu, Zhi Gao","submitted_at":"2024-12-20T07:00:46Z","abstract_excerpt":"The advancement of large language models (LLMs) prompts the development of multi-modal agents, which are used as a controller to call external tools, providing a feasible way to solve practical tasks. In this paper, we propose a multi-modal agent tuning method that automatically generates multi-modal tool-usage data and tunes a vision-language model (VLM) as the controller for powerful tool-usage reasoning. To preserve the data quality, we prompt the GPT-4o mini model to generate queries, files, and trajectories, followed by query-file and trajectory verifiers. Based on the data synthesis pipe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.15606","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/2412.15606/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:08:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"epVu+u9vuQMXPuafyUgO64GOQpVxBVVHlIzGS74KlL4nMc8xwbbfIsF/vEkuqETm3bvgAZB9VpqSzwcGwTvyAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T17:55:04.863230Z"},"content_sha256":"04b13ba71b203fb3210d86221d4d96f444209f2ce3c45b4d2b0ec1030cdacf63","schema_version":"1.0","event_id":"sha256:04b13ba71b203fb3210d86221d4d96f444209f2ce3c45b4d2b0ec1030cdacf63"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4LPIUM4K4ZSICKQY6IRO7PWX4Z/bundle.json","state_url":"https://pith.science/pith/4LPIUM4K4ZSICKQY6IRO7PWX4Z/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4LPIUM4K4ZSICKQY6IRO7PWX4Z/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-08T17:55:04Z","links":{"resolver":"https://pith.science/pith/4LPIUM4K4ZSICKQY6IRO7PWX4Z","bundle":"https://pith.science/pith/4LPIUM4K4ZSICKQY6IRO7PWX4Z/bundle.json","state":"https://pith.science/pith/4LPIUM4K4ZSICKQY6IRO7PWX4Z/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4LPIUM4K4ZSICKQY6IRO7PWX4Z/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:4LPIUM4K4ZSICKQY6IRO7PWX4Z","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"6a780f055d99d88f21a3742af55e51ef83c9b5ec17d346d4109e5536c2d6c10b","cross_cats_sorted":["cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-12-20T07:00:46Z","title_canon_sha256":"4258df22b44040723a5bb8c47a5789ff9bc51b62bc6fbc449fb4a20c9e4ae093"},"schema_version":"1.0","source":{"id":"2412.15606","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.15606","created_at":"2026-07-05T10:08:26Z"},{"alias_kind":"arxiv_version","alias_value":"2412.15606v2","created_at":"2026-07-05T10:08:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.15606","created_at":"2026-07-05T10:08:26Z"},{"alias_kind":"pith_short_12","alias_value":"4LPIUM4K4ZSI","created_at":"2026-07-05T10:08:26Z"},{"alias_kind":"pith_short_16","alias_value":"4LPIUM4K4ZSICKQY","created_at":"2026-07-05T10:08:26Z"},{"alias_kind":"pith_short_8","alias_value":"4LPIUM4K","created_at":"2026-07-05T10:08:26Z"}],"graph_snapshots":[{"event_id":"sha256:04b13ba71b203fb3210d86221d4d96f444209f2ce3c45b4d2b0ec1030cdacf63","target":"graph","created_at":"2026-07-05T10:08:26Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2412.15606/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The advancement of large language models (LLMs) prompts the development of multi-modal agents, which are used as a controller to call external tools, providing a feasible way to solve practical tasks. In this paper, we propose a multi-modal agent tuning method that automatically generates multi-modal tool-usage data and tunes a vision-language model (VLM) as the controller for powerful tool-usage reasoning. To preserve the data quality, we prompt the GPT-4o mini model to generate queries, files, and trajectories, followed by query-file and trajectory verifiers. Based on the data synthesis pipe","authors_text":"Bofei Zhang, Pengxiang Li, Qing Li, Song-Chun Zhu, Tao Yuan, Xiaojian Ma, Yue Fan, Yunde Jia, Yuwei Wu, Zhi Gao","cross_cats":["cs.CV"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-12-20T07:00:46Z","title":"Multi-modal Agent Tuning: Building a VLM-Driven Agent for Efficient Tool Usage"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.15606","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:324306679d0a9efa41fae88d6c86eb5bc721e4ab1500f7ee955ebfe23cf3b9b7","target":"record","created_at":"2026-07-05T10:08:26Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"6a780f055d99d88f21a3742af55e51ef83c9b5ec17d346d4109e5536c2d6c10b","cross_cats_sorted":["cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-12-20T07:00:46Z","title_canon_sha256":"4258df22b44040723a5bb8c47a5789ff9bc51b62bc6fbc449fb4a20c9e4ae093"},"schema_version":"1.0","source":{"id":"2412.15606","kind":"arxiv","version":2}},"canonical_sha256":"e2de8a338ae664812a18f222efbed7e669689ce6d9807d8e607b9d03afae0760","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e2de8a338ae664812a18f222efbed7e669689ce6d9807d8e607b9d03afae0760","first_computed_at":"2026-07-05T10:08:26.605350Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:08:26.605350Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mptRWRE+Q643XjssjspKIsRwWc+CbSheuK4ls60mRBvq3DmVAMc5bYSZb3obot0uSXOOHZfEKgmPqyTFnImqBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:08:26.605843Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.15606","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:324306679d0a9efa41fae88d6c86eb5bc721e4ab1500f7ee955ebfe23cf3b9b7","sha256:04b13ba71b203fb3210d86221d4d96f444209f2ce3c45b4d2b0ec1030cdacf63"],"state_sha256":"1a4446eb00d6de50404f79bd600dd68c974c7aad52596b0615e2d3140fd88c9b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+o6m54PbE4g6d9ChEC7q8l+CwnXE+Z+RE8mTbLsXcj7HKkzK2OfMMQ/kgCp7P50b0ZUjr+Ir4c5gtc/7PCmHBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T17:55:04.866744Z","bundle_sha256":"00b7d78ddc02ef9cfcd696e49c7136bd8919e36cba838455df0040d623e132ee"}}