{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:OXT23RBEMNZC4DCQU2MMLOHLRH","short_pith_number":"pith:OXT23RBE","schema_version":"1.0","canonical_sha256":"75e7adc42463722e0c50a698c5b8eb89f4936579671beaac6de686adf4a0fb94","source":{"kind":"arxiv","id":"2506.19848","version":1},"attestation_state":"computed","paper":{"title":"ScaleCap: Inference-Time Scalable Image Captioning via Dual-Modality Debiasing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Dahua Lin, Feng Wu, Jiaqi Wang, Jinsong Li, Long Xing, Nenghai Yu, Pan Zhang, Qidong Huang, Shuangrui Ding, Weiming Zhang, Xiaoyi Dong, Yuhang Cao, Yuhang Zang","submitted_at":"2025-06-24T17:59:55Z","abstract_excerpt":"This paper presents ScaleCap, an inference-time scalable image captioning strategy that generates comprehensive and detailed image captions. The key challenges of high-quality image captioning lie in the inherent biases of LVLMs: multimodal bias resulting in imbalanced descriptive granularity, offering detailed accounts of some elements while merely skimming over others; linguistic bias leading to hallucinated descriptions of non-existent objects. To address these issues, we propose a scalable debiased captioning strategy, which continuously enriches and calibrates the caption with increased i"},"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":"2506.19848","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-06-24T17:59:55Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"d869ebb27eb8c1c069a78fc293a20423a3026b95d269bc534dfd2d9170fce9ed","abstract_canon_sha256":"1b280141c562fd5dfd40ce775595bf20868c876cc08983b4a31c4a0cf6bc0087"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:26:39.512898Z","signature_b64":"yxIRn/Q6diO4rY7hDI2WkTQ2vnbRyTA7qwUOJ7rdBUptov/4rqG1ZjdZja3d1A6iDRE3WtH1S2luyTPL1lnaDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"75e7adc42463722e0c50a698c5b8eb89f4936579671beaac6de686adf4a0fb94","last_reissued_at":"2026-07-05T11:26:39.512416Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:26:39.512416Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ScaleCap: Inference-Time Scalable Image Captioning via Dual-Modality Debiasing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Dahua Lin, Feng Wu, Jiaqi Wang, Jinsong Li, Long Xing, Nenghai Yu, Pan Zhang, Qidong Huang, Shuangrui Ding, Weiming Zhang, Xiaoyi Dong, Yuhang Cao, Yuhang Zang","submitted_at":"2025-06-24T17:59:55Z","abstract_excerpt":"This paper presents ScaleCap, an inference-time scalable image captioning strategy that generates comprehensive and detailed image captions. The key challenges of high-quality image captioning lie in the inherent biases of LVLMs: multimodal bias resulting in imbalanced descriptive granularity, offering detailed accounts of some elements while merely skimming over others; linguistic bias leading to hallucinated descriptions of non-existent objects. To address these issues, we propose a scalable debiased captioning strategy, which continuously enriches and calibrates the caption with increased i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.19848","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/2506.19848/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":"2506.19848","created_at":"2026-07-05T11:26:39.512471+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.19848v1","created_at":"2026-07-05T11:26:39.512471+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.19848","created_at":"2026-07-05T11:26:39.512471+00:00"},{"alias_kind":"pith_short_12","alias_value":"OXT23RBEMNZC","created_at":"2026-07-05T11:26:39.512471+00:00"},{"alias_kind":"pith_short_16","alias_value":"OXT23RBEMNZC4DCQ","created_at":"2026-07-05T11:26:39.512471+00:00"},{"alias_kind":"pith_short_8","alias_value":"OXT23RBE","created_at":"2026-07-05T11:26:39.512471+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.05623","citing_title":"DetailVerifyBench: A Benchmark for Dense Hallucination Localization in Long Image Captions","ref_index":49,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OXT23RBEMNZC4DCQU2MMLOHLRH","json":"https://pith.science/pith/OXT23RBEMNZC4DCQU2MMLOHLRH.json","graph_json":"https://pith.science/api/pith-number/OXT23RBEMNZC4DCQU2MMLOHLRH/graph.json","events_json":"https://pith.science/api/pith-number/OXT23RBEMNZC4DCQU2MMLOHLRH/events.json","paper":"https://pith.science/paper/OXT23RBE"},"agent_actions":{"view_html":"https://pith.science/pith/OXT23RBEMNZC4DCQU2MMLOHLRH","download_json":"https://pith.science/pith/OXT23RBEMNZC4DCQU2MMLOHLRH.json","view_paper":"https://pith.science/paper/OXT23RBE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.19848&json=true","fetch_graph":"https://pith.science/api/pith-number/OXT23RBEMNZC4DCQU2MMLOHLRH/graph.json","fetch_events":"https://pith.science/api/pith-number/OXT23RBEMNZC4DCQU2MMLOHLRH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OXT23RBEMNZC4DCQU2MMLOHLRH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OXT23RBEMNZC4DCQU2MMLOHLRH/action/storage_attestation","attest_author":"https://pith.science/pith/OXT23RBEMNZC4DCQU2MMLOHLRH/action/author_attestation","sign_citation":"https://pith.science/pith/OXT23RBEMNZC4DCQU2MMLOHLRH/action/citation_signature","submit_replication":"https://pith.science/pith/OXT23RBEMNZC4DCQU2MMLOHLRH/action/replication_record"}},"created_at":"2026-07-05T11:26:39.512471+00:00","updated_at":"2026-07-05T11:26:39.512471+00:00"}