{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:276DMUK55H54GYKT4MPJSVXPFZ","short_pith_number":"pith:276DMUK5","schema_version":"1.0","canonical_sha256":"d7fc36515de9fbc36153e31e9956ef2e78db7244aa35dbf782ae7bc428825ff2","source":{"kind":"arxiv","id":"2510.10921","version":3},"attestation_state":"computed","paper":{"title":"FG-CLIP 2: A Bilingual Fine-grained Vision-Language Alignment Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Bin Wang, Chunyu Xie, Dawei Leng, Dawei Liang, Fanjing Kong, Ji Ao, Jincheng Li, Yuhui Yin","submitted_at":"2025-10-13T02:32:07Z","abstract_excerpt":"Fine-grained vision-language understanding requires precise alignment between visual content and linguistic descriptions, a capability that remains limited in current models, particularly in non-English settings. While models like CLIP perform well on global alignment, they often struggle to capture fine-grained details in object attributes, spatial relations, and linguistic expressions, with limited support for bilingual comprehension. To address these challenges, we introduce FG-CLIP 2, a bilingual vision-language model designed to advance fine-grained alignment for both English and Chinese."},"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":"2510.10921","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-10-13T02:32:07Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"dee1ffa2278edf7ea577f76c39316a9c829d3e1753975940e14260d09f3d5bae","abstract_canon_sha256":"069f293147bc31364e5b293bf9bf35cc363645e1abfd7fdac33d18ac574c8844"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T02:03:58.027570Z","signature_b64":"7IHBOkoVXAYsUNbVEgampw6hkQ4j2eU/8R4K6B8QDl5xo+5eRvhyX/R/uMrrGseIzjzEYFxtIeFP0R9VEMyrCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d7fc36515de9fbc36153e31e9956ef2e78db7244aa35dbf782ae7bc428825ff2","last_reissued_at":"2026-05-26T02:03:58.026691Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T02:03:58.026691Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FG-CLIP 2: A Bilingual Fine-grained Vision-Language Alignment Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Bin Wang, Chunyu Xie, Dawei Leng, Dawei Liang, Fanjing Kong, Ji Ao, Jincheng Li, Yuhui Yin","submitted_at":"2025-10-13T02:32:07Z","abstract_excerpt":"Fine-grained vision-language understanding requires precise alignment between visual content and linguistic descriptions, a capability that remains limited in current models, particularly in non-English settings. While models like CLIP perform well on global alignment, they often struggle to capture fine-grained details in object attributes, spatial relations, and linguistic expressions, with limited support for bilingual comprehension. To address these challenges, we introduce FG-CLIP 2, a bilingual vision-language model designed to advance fine-grained alignment for both English and Chinese."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.10921","kind":"arxiv","version":3},"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/2510.10921/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":"2510.10921","created_at":"2026-05-26T02:03:58.026822+00:00"},{"alias_kind":"arxiv_version","alias_value":"2510.10921v3","created_at":"2026-05-26T02:03:58.026822+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.10921","created_at":"2026-05-26T02:03:58.026822+00:00"},{"alias_kind":"pith_short_12","alias_value":"276DMUK55H54","created_at":"2026-05-26T02:03:58.026822+00:00"},{"alias_kind":"pith_short_16","alias_value":"276DMUK55H54GYKT","created_at":"2026-05-26T02:03:58.026822+00:00"},{"alias_kind":"pith_short_8","alias_value":"276DMUK5","created_at":"2026-05-26T02:03:58.026822+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2605.20147","citing_title":"PixVerve: Advancing Native UHR Image Generation to 100MP with a Large-Scale High-Quality Dataset","ref_index":51,"is_internal_anchor":true},{"citing_arxiv_id":"2604.22855","citing_title":"Evaluating Remote Sensing Image Captions Beyond Metric Biases","ref_index":55,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/276DMUK55H54GYKT4MPJSVXPFZ","json":"https://pith.science/pith/276DMUK55H54GYKT4MPJSVXPFZ.json","graph_json":"https://pith.science/api/pith-number/276DMUK55H54GYKT4MPJSVXPFZ/graph.json","events_json":"https://pith.science/api/pith-number/276DMUK55H54GYKT4MPJSVXPFZ/events.json","paper":"https://pith.science/paper/276DMUK5"},"agent_actions":{"view_html":"https://pith.science/pith/276DMUK55H54GYKT4MPJSVXPFZ","download_json":"https://pith.science/pith/276DMUK55H54GYKT4MPJSVXPFZ.json","view_paper":"https://pith.science/paper/276DMUK5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2510.10921&json=true","fetch_graph":"https://pith.science/api/pith-number/276DMUK55H54GYKT4MPJSVXPFZ/graph.json","fetch_events":"https://pith.science/api/pith-number/276DMUK55H54GYKT4MPJSVXPFZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/276DMUK55H54GYKT4MPJSVXPFZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/276DMUK55H54GYKT4MPJSVXPFZ/action/storage_attestation","attest_author":"https://pith.science/pith/276DMUK55H54GYKT4MPJSVXPFZ/action/author_attestation","sign_citation":"https://pith.science/pith/276DMUK55H54GYKT4MPJSVXPFZ/action/citation_signature","submit_replication":"https://pith.science/pith/276DMUK55H54GYKT4MPJSVXPFZ/action/replication_record"}},"created_at":"2026-05-26T02:03:58.026822+00:00","updated_at":"2026-05-26T02:03:58.026822+00:00"}