{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ZGE5SN27VM4KYCXBJ4LLTW3N6E","short_pith_number":"pith:ZGE5SN27","schema_version":"1.0","canonical_sha256":"c989d9375fab38ac0ae14f16b9db6df131fbc7d7f9d44f41835a830cff67808c","source":{"kind":"arxiv","id":"2509.03897","version":2},"attestation_state":"computed","paper":{"title":"SPECS: Specificity-Enhanced CLIP-Score for Long Image Caption Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Israfel Salazar, Xiaofu Chen, Yova Kementchedjhieva","submitted_at":"2025-09-04T05:43:50Z","abstract_excerpt":"As interest grows in generating long, detailed image captions, standard evaluation metrics become increasingly unreliable. N-gram-based metrics though efficient, fail to capture semantic correctness. Representational Similarity (RS) metrics, designed to address this, initially saw limited use due to high computational costs, while today, despite advances in hardware, they remain unpopular due to low correlation to human judgments. Meanwhile, metrics based on large language models (LLMs) show strong correlation with human judgments, but remain too expensive for iterative use during model develo"},"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":"2509.03897","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-09-04T05:43:50Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"22b1af618917c4f00ecf40194eb3af7d8be5423787ac2dfa231d8d26704856fc","abstract_canon_sha256":"e785bf3e8bbadced778d4607294d7ae3e2211cebc0cab6cb5da77e47c3564d86"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:09:46.197277Z","signature_b64":"ULDHH5Mbu23T2CrQ95iDM7t4/r4calVFN4LaKTwqS5fCD08dquvCcR+c72bszlGmYKsQ+w7L0ubbnVVnbU0oCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c989d9375fab38ac0ae14f16b9db6df131fbc7d7f9d44f41835a830cff67808c","last_reissued_at":"2026-07-05T12:09:46.196706Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:09:46.196706Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SPECS: Specificity-Enhanced CLIP-Score for Long Image Caption Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Israfel Salazar, Xiaofu Chen, Yova Kementchedjhieva","submitted_at":"2025-09-04T05:43:50Z","abstract_excerpt":"As interest grows in generating long, detailed image captions, standard evaluation metrics become increasingly unreliable. N-gram-based metrics though efficient, fail to capture semantic correctness. Representational Similarity (RS) metrics, designed to address this, initially saw limited use due to high computational costs, while today, despite advances in hardware, they remain unpopular due to low correlation to human judgments. Meanwhile, metrics based on large language models (LLMs) show strong correlation with human judgments, but remain too expensive for iterative use during model develo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.03897","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/2509.03897/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":"2509.03897","created_at":"2026-07-05T12:09:46.196768+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.03897v2","created_at":"2026-07-05T12:09:46.196768+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.03897","created_at":"2026-07-05T12:09:46.196768+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZGE5SN27VM4K","created_at":"2026-07-05T12:09:46.196768+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZGE5SN27VM4KYCXB","created_at":"2026-07-05T12:09:46.196768+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZGE5SN27","created_at":"2026-07-05T12:09:46.196768+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/ZGE5SN27VM4KYCXBJ4LLTW3N6E","json":"https://pith.science/pith/ZGE5SN27VM4KYCXBJ4LLTW3N6E.json","graph_json":"https://pith.science/api/pith-number/ZGE5SN27VM4KYCXBJ4LLTW3N6E/graph.json","events_json":"https://pith.science/api/pith-number/ZGE5SN27VM4KYCXBJ4LLTW3N6E/events.json","paper":"https://pith.science/paper/ZGE5SN27"},"agent_actions":{"view_html":"https://pith.science/pith/ZGE5SN27VM4KYCXBJ4LLTW3N6E","download_json":"https://pith.science/pith/ZGE5SN27VM4KYCXBJ4LLTW3N6E.json","view_paper":"https://pith.science/paper/ZGE5SN27","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.03897&json=true","fetch_graph":"https://pith.science/api/pith-number/ZGE5SN27VM4KYCXBJ4LLTW3N6E/graph.json","fetch_events":"https://pith.science/api/pith-number/ZGE5SN27VM4KYCXBJ4LLTW3N6E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZGE5SN27VM4KYCXBJ4LLTW3N6E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZGE5SN27VM4KYCXBJ4LLTW3N6E/action/storage_attestation","attest_author":"https://pith.science/pith/ZGE5SN27VM4KYCXBJ4LLTW3N6E/action/author_attestation","sign_citation":"https://pith.science/pith/ZGE5SN27VM4KYCXBJ4LLTW3N6E/action/citation_signature","submit_replication":"https://pith.science/pith/ZGE5SN27VM4KYCXBJ4LLTW3N6E/action/replication_record"}},"created_at":"2026-07-05T12:09:46.196768+00:00","updated_at":"2026-07-05T12:09:46.196768+00:00"}