{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:EN5IBSMSXQS2ZGUIULAW4POB2K","short_pith_number":"pith:EN5IBSMS","schema_version":"1.0","canonical_sha256":"237a80c992bc25ac9a88a2c16e3dc1d2a2a7b1cedbee9f502b297c5afe824540","source":{"kind":"arxiv","id":"2504.06666","version":1},"attestation_state":"computed","paper":{"title":"Patch Matters: Training-free Fine-grained Image Caption Enhancement via Local Perception","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Di Hu, Haiying He, Ruotian Peng, Yake Wei, Yandong Wen","submitted_at":"2025-04-09T08:07:46Z","abstract_excerpt":"High-quality image captions play a crucial role in improving the performance of cross-modal applications such as text-to-image generation, text-to-video generation, and text-image retrieval. To generate long-form, high-quality captions, many recent studies have employed multimodal large language models (MLLMs). However, current MLLMs often produce captions that lack fine-grained details or suffer from hallucinations, a challenge that persists in both open-source and closed-source models. Inspired by Feature-Integration theory, which suggests that attention must focus on specific regions to int"},"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":"2504.06666","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-04-09T08:07:46Z","cross_cats_sorted":[],"title_canon_sha256":"5e8ff22d103f0767ff2f3c7ca82dbd39afd774bbe3c8651e1205a0dbee916794","abstract_canon_sha256":"791ccb460b8f89419d283f4ed3337e634f0ba4e5eb947194218015c09d3a1557"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:46:43.782906Z","signature_b64":"6t/YxMemXPsoDJ4x5jMOY3b3TLabJnok0QAe1v527wlQAGHU59NlDXCASkcuuDX4a6Mlv0MUp5gQU/HKOC6GBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"237a80c992bc25ac9a88a2c16e3dc1d2a2a7b1cedbee9f502b297c5afe824540","last_reissued_at":"2026-07-05T10:46:43.782397Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:46:43.782397Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Patch Matters: Training-free Fine-grained Image Caption Enhancement via Local Perception","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Di Hu, Haiying He, Ruotian Peng, Yake Wei, Yandong Wen","submitted_at":"2025-04-09T08:07:46Z","abstract_excerpt":"High-quality image captions play a crucial role in improving the performance of cross-modal applications such as text-to-image generation, text-to-video generation, and text-image retrieval. To generate long-form, high-quality captions, many recent studies have employed multimodal large language models (MLLMs). However, current MLLMs often produce captions that lack fine-grained details or suffer from hallucinations, a challenge that persists in both open-source and closed-source models. Inspired by Feature-Integration theory, which suggests that attention must focus on specific regions to int"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.06666","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/2504.06666/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":"2504.06666","created_at":"2026-07-05T10:46:43.782462+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.06666v1","created_at":"2026-07-05T10:46:43.782462+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.06666","created_at":"2026-07-05T10:46:43.782462+00:00"},{"alias_kind":"pith_short_12","alias_value":"EN5IBSMSXQS2","created_at":"2026-07-05T10:46:43.782462+00:00"},{"alias_kind":"pith_short_16","alias_value":"EN5IBSMSXQS2ZGUI","created_at":"2026-07-05T10:46:43.782462+00:00"},{"alias_kind":"pith_short_8","alias_value":"EN5IBSMS","created_at":"2026-07-05T10:46:43.782462+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.22613","citing_title":"RICO: Improving Accuracy and Completeness in Image Recaptioning via Visual Reconstruction","ref_index":37,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EN5IBSMSXQS2ZGUIULAW4POB2K","json":"https://pith.science/pith/EN5IBSMSXQS2ZGUIULAW4POB2K.json","graph_json":"https://pith.science/api/pith-number/EN5IBSMSXQS2ZGUIULAW4POB2K/graph.json","events_json":"https://pith.science/api/pith-number/EN5IBSMSXQS2ZGUIULAW4POB2K/events.json","paper":"https://pith.science/paper/EN5IBSMS"},"agent_actions":{"view_html":"https://pith.science/pith/EN5IBSMSXQS2ZGUIULAW4POB2K","download_json":"https://pith.science/pith/EN5IBSMSXQS2ZGUIULAW4POB2K.json","view_paper":"https://pith.science/paper/EN5IBSMS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.06666&json=true","fetch_graph":"https://pith.science/api/pith-number/EN5IBSMSXQS2ZGUIULAW4POB2K/graph.json","fetch_events":"https://pith.science/api/pith-number/EN5IBSMSXQS2ZGUIULAW4POB2K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EN5IBSMSXQS2ZGUIULAW4POB2K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EN5IBSMSXQS2ZGUIULAW4POB2K/action/storage_attestation","attest_author":"https://pith.science/pith/EN5IBSMSXQS2ZGUIULAW4POB2K/action/author_attestation","sign_citation":"https://pith.science/pith/EN5IBSMSXQS2ZGUIULAW4POB2K/action/citation_signature","submit_replication":"https://pith.science/pith/EN5IBSMSXQS2ZGUIULAW4POB2K/action/replication_record"}},"created_at":"2026-07-05T10:46:43.782462+00:00","updated_at":"2026-07-05T10:46:43.782462+00:00"}