{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:4ORJXWLST3KG5QIA72O4NFZQEW","short_pith_number":"pith:4ORJXWLS","schema_version":"1.0","canonical_sha256":"e3a29bd9729ed46ec100fe9dc6973025b76b771e81f75e006feb928f7009bb63","source":{"kind":"arxiv","id":"2606.07645","version":1},"attestation_state":"computed","paper":{"title":"FineGen: A VLM-based Multi-Agent Framework for Fine-Grained Image-Text Dataset Construction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chang Kong, Haigang Zhang, Peng Mo, Qiuming Luo, Yuebing Li","submitted_at":"2026-06-02T01:10:12Z","abstract_excerpt":"The scarcity of hard negative samples in current vision-language datasets significantly hinders fine-grained perception. To address this, we propose FineGen, a VLM-based Multi-Agent framework for automated dataset construction. By employing a collaborative Generation-Verification-Correction pipeline with a closed-loop feedback mechanism, FineGen ensures synthesized hard negatives are semantically valid yet strictly contradictory to visual content. Applying this to ImageNet, we construct FineGen-100K, a hierarchical dataset containing over 147,000 attribute-specific hard negatives with a rigoro"},"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":"2606.07645","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-02T01:10:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f843714cc83bcb4db1a85ebc98b07f0be8ff2b7996851c1f5b3c8391f5f455cf","abstract_canon_sha256":"602588dd00c5e3822672bee285d9d0a7b87852625d28ee58f95da388aab9ef77"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T00:04:45.704500Z","signature_b64":"VF9dwC5DwnAGieSLsY7DIk6PWOf5DElveDsha8ASkyvDTUk9T6qHMf2qUNYoytj6RKar9Q/QwZkW+/llJF+sAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e3a29bd9729ed46ec100fe9dc6973025b76b771e81f75e006feb928f7009bb63","last_reissued_at":"2026-06-09T00:04:45.704139Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T00:04:45.704139Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FineGen: A VLM-based Multi-Agent Framework for Fine-Grained Image-Text Dataset Construction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chang Kong, Haigang Zhang, Peng Mo, Qiuming Luo, Yuebing Li","submitted_at":"2026-06-02T01:10:12Z","abstract_excerpt":"The scarcity of hard negative samples in current vision-language datasets significantly hinders fine-grained perception. To address this, we propose FineGen, a VLM-based Multi-Agent framework for automated dataset construction. By employing a collaborative Generation-Verification-Correction pipeline with a closed-loop feedback mechanism, FineGen ensures synthesized hard negatives are semantically valid yet strictly contradictory to visual content. Applying this to ImageNet, we construct FineGen-100K, a hierarchical dataset containing over 147,000 attribute-specific hard negatives with a rigoro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.07645","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/2606.07645/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":"2606.07645","created_at":"2026-06-09T00:04:45.704198+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.07645v1","created_at":"2026-06-09T00:04:45.704198+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.07645","created_at":"2026-06-09T00:04:45.704198+00:00"},{"alias_kind":"pith_short_12","alias_value":"4ORJXWLST3KG","created_at":"2026-06-09T00:04:45.704198+00:00"},{"alias_kind":"pith_short_16","alias_value":"4ORJXWLST3KG5QIA","created_at":"2026-06-09T00:04:45.704198+00:00"},{"alias_kind":"pith_short_8","alias_value":"4ORJXWLS","created_at":"2026-06-09T00:04:45.704198+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/4ORJXWLST3KG5QIA72O4NFZQEW","json":"https://pith.science/pith/4ORJXWLST3KG5QIA72O4NFZQEW.json","graph_json":"https://pith.science/api/pith-number/4ORJXWLST3KG5QIA72O4NFZQEW/graph.json","events_json":"https://pith.science/api/pith-number/4ORJXWLST3KG5QIA72O4NFZQEW/events.json","paper":"https://pith.science/paper/4ORJXWLS"},"agent_actions":{"view_html":"https://pith.science/pith/4ORJXWLST3KG5QIA72O4NFZQEW","download_json":"https://pith.science/pith/4ORJXWLST3KG5QIA72O4NFZQEW.json","view_paper":"https://pith.science/paper/4ORJXWLS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.07645&json=true","fetch_graph":"https://pith.science/api/pith-number/4ORJXWLST3KG5QIA72O4NFZQEW/graph.json","fetch_events":"https://pith.science/api/pith-number/4ORJXWLST3KG5QIA72O4NFZQEW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4ORJXWLST3KG5QIA72O4NFZQEW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4ORJXWLST3KG5QIA72O4NFZQEW/action/storage_attestation","attest_author":"https://pith.science/pith/4ORJXWLST3KG5QIA72O4NFZQEW/action/author_attestation","sign_citation":"https://pith.science/pith/4ORJXWLST3KG5QIA72O4NFZQEW/action/citation_signature","submit_replication":"https://pith.science/pith/4ORJXWLST3KG5QIA72O4NFZQEW/action/replication_record"}},"created_at":"2026-06-09T00:04:45.704198+00:00","updated_at":"2026-06-09T00:04:45.704198+00:00"}