{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:5IWJHZDB2WF56MSII3B53ZFBC5","short_pith_number":"pith:5IWJHZDB","schema_version":"1.0","canonical_sha256":"ea2c93e461d58bdf324846c3dde4a1174bdcfa1199365f2dbe68cc428740b1e4","source":{"kind":"arxiv","id":"2409.13540","version":1},"attestation_state":"computed","paper":{"title":"FullAnno: A Data Engine for Enhancing Image Comprehension of MLLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Errui Ding, Gang Zhang, Jingdong Wang, Jing Hao, Kun Yao, Qiang Chen, Song Chen, Yanpeng Sun, Yuxiang Zhao","submitted_at":"2024-09-20T14:33:17Z","abstract_excerpt":"Multimodal Large Language Models (MLLMs) have shown promise in a broad range of vision-language tasks with their strong reasoning and generalization capabilities. However, they heavily depend on high-quality data in the Supervised Fine-Tuning (SFT) phase. The existing approaches aim to curate high-quality data via GPT-4V, but they are not scalable due to the commercial nature of GPT-4V and the simplicity of the prompts used to instruct the model. To this end, we devised the FullAnno system, which is a data engine that can generate large-scale, high-quality, and fine-grained image annotations c"},"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":"2409.13540","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-09-20T14:33:17Z","cross_cats_sorted":[],"title_canon_sha256":"a0ce9042ab08f6f935e151adf523e2ad91b7facb87b669ce25a50c70aa5f89ad","abstract_canon_sha256":"ad90150fd7697afd6fd67c188c6c6aa0b58cb8b2d5f8e6339cfa794889fdd77e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:09:37.287234Z","signature_b64":"8YT6aWanxqxlvpXMUTzadMQTdpCHvCnP5TW60MPXtU2Wsn/khXoTTS7KxeQcsJ+yGBo9RCnLSNkQRD2i7YT7Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ea2c93e461d58bdf324846c3dde4a1174bdcfa1199365f2dbe68cc428740b1e4","last_reissued_at":"2026-07-05T09:09:37.286720Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:09:37.286720Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FullAnno: A Data Engine for Enhancing Image Comprehension of MLLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Errui Ding, Gang Zhang, Jingdong Wang, Jing Hao, Kun Yao, Qiang Chen, Song Chen, Yanpeng Sun, Yuxiang Zhao","submitted_at":"2024-09-20T14:33:17Z","abstract_excerpt":"Multimodal Large Language Models (MLLMs) have shown promise in a broad range of vision-language tasks with their strong reasoning and generalization capabilities. However, they heavily depend on high-quality data in the Supervised Fine-Tuning (SFT) phase. The existing approaches aim to curate high-quality data via GPT-4V, but they are not scalable due to the commercial nature of GPT-4V and the simplicity of the prompts used to instruct the model. To this end, we devised the FullAnno system, which is a data engine that can generate large-scale, high-quality, and fine-grained image annotations c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.13540","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/2409.13540/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":"2409.13540","created_at":"2026-07-05T09:09:37.286775+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.13540v1","created_at":"2026-07-05T09:09:37.286775+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.13540","created_at":"2026-07-05T09:09:37.286775+00:00"},{"alias_kind":"pith_short_12","alias_value":"5IWJHZDB2WF5","created_at":"2026-07-05T09:09:37.286775+00:00"},{"alias_kind":"pith_short_16","alias_value":"5IWJHZDB2WF56MSI","created_at":"2026-07-05T09:09:37.286775+00:00"},{"alias_kind":"pith_short_8","alias_value":"5IWJHZDB","created_at":"2026-07-05T09:09:37.286775+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.17288","citing_title":"When Efficiency Backfires: Cascading LLMs Trigger Cascade Failure under Adversarial Attack","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2412.05579","citing_title":"LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methods","ref_index":82,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5IWJHZDB2WF56MSII3B53ZFBC5","json":"https://pith.science/pith/5IWJHZDB2WF56MSII3B53ZFBC5.json","graph_json":"https://pith.science/api/pith-number/5IWJHZDB2WF56MSII3B53ZFBC5/graph.json","events_json":"https://pith.science/api/pith-number/5IWJHZDB2WF56MSII3B53ZFBC5/events.json","paper":"https://pith.science/paper/5IWJHZDB"},"agent_actions":{"view_html":"https://pith.science/pith/5IWJHZDB2WF56MSII3B53ZFBC5","download_json":"https://pith.science/pith/5IWJHZDB2WF56MSII3B53ZFBC5.json","view_paper":"https://pith.science/paper/5IWJHZDB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.13540&json=true","fetch_graph":"https://pith.science/api/pith-number/5IWJHZDB2WF56MSII3B53ZFBC5/graph.json","fetch_events":"https://pith.science/api/pith-number/5IWJHZDB2WF56MSII3B53ZFBC5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5IWJHZDB2WF56MSII3B53ZFBC5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5IWJHZDB2WF56MSII3B53ZFBC5/action/storage_attestation","attest_author":"https://pith.science/pith/5IWJHZDB2WF56MSII3B53ZFBC5/action/author_attestation","sign_citation":"https://pith.science/pith/5IWJHZDB2WF56MSII3B53ZFBC5/action/citation_signature","submit_replication":"https://pith.science/pith/5IWJHZDB2WF56MSII3B53ZFBC5/action/replication_record"}},"created_at":"2026-07-05T09:09:37.286775+00:00","updated_at":"2026-07-05T09:09:37.286775+00:00"}