{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HGO6IF7RM3RSRSDOVHT7WA5NCQ","short_pith_number":"pith:HGO6IF7R","schema_version":"1.0","canonical_sha256":"399de417f166e328c86ea9e7fb03ad141f2575f720964fbf4bb990395ab5e252","source":{"kind":"arxiv","id":"2503.12329","version":1},"attestation_state":"computed","paper":{"title":"CapArena: Benchmarking and Analyzing Detailed Image Captioning in the LLM Era","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Chenyang Yan, Fangzhi Xu, Jiajun Chen, Jianbing Zhang, Jiaxin Fan, Kanzhi Cheng, Nuo Chen, Qiushi Sun, Wenpo Song, Zheng Ma","submitted_at":"2025-03-16T02:56:09Z","abstract_excerpt":"Image captioning has been a longstanding challenge in vision-language research. With the rise of LLMs, modern Vision-Language Models (VLMs) generate detailed and comprehensive image descriptions. However, benchmarking the quality of such captions remains unresolved. This paper addresses two key questions: (1) How well do current VLMs actually perform on image captioning, particularly compared to humans? We built CapArena, a platform with over 6000 pairwise caption battles and high-quality human preference votes. Our arena-style evaluation marks a milestone, showing that leading models like GPT"},"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":"2503.12329","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-16T02:56:09Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"bd9480a19cbb5a7d6190af5be0512bc0e3692243457009a9fbe3d4d851c43c88","abstract_canon_sha256":"8871315aee376cfa1dc7018805cd160ac6c0b056d85780f0c522da9b0ecb9445"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:32:10.613702Z","signature_b64":"S3tWpoDaCg1zF656PvT0rtrT2OW4mcf/J7qFleVmEHTqguRxy74PrdA1TdklS3QjD572DtrHVjTk79HrCVlJBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"399de417f166e328c86ea9e7fb03ad141f2575f720964fbf4bb990395ab5e252","last_reissued_at":"2026-07-05T10:32:10.612711Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:32:10.612711Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CapArena: Benchmarking and Analyzing Detailed Image Captioning in the LLM Era","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Chenyang Yan, Fangzhi Xu, Jiajun Chen, Jianbing Zhang, Jiaxin Fan, Kanzhi Cheng, Nuo Chen, Qiushi Sun, Wenpo Song, Zheng Ma","submitted_at":"2025-03-16T02:56:09Z","abstract_excerpt":"Image captioning has been a longstanding challenge in vision-language research. With the rise of LLMs, modern Vision-Language Models (VLMs) generate detailed and comprehensive image descriptions. However, benchmarking the quality of such captions remains unresolved. This paper addresses two key questions: (1) How well do current VLMs actually perform on image captioning, particularly compared to humans? We built CapArena, a platform with over 6000 pairwise caption battles and high-quality human preference votes. Our arena-style evaluation marks a milestone, showing that leading models like GPT"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.12329","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/2503.12329/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":"2503.12329","created_at":"2026-07-05T10:32:10.612882+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.12329v1","created_at":"2026-07-05T10:32:10.612882+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.12329","created_at":"2026-07-05T10:32:10.612882+00:00"},{"alias_kind":"pith_short_12","alias_value":"HGO6IF7RM3RS","created_at":"2026-07-05T10:32:10.612882+00:00"},{"alias_kind":"pith_short_16","alias_value":"HGO6IF7RM3RSRSDO","created_at":"2026-07-05T10:32:10.612882+00:00"},{"alias_kind":"pith_short_8","alias_value":"HGO6IF7R","created_at":"2026-07-05T10:32:10.612882+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26734","citing_title":"Robust Onion: Peeling Open Vocab Object Detectors Under Noise","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05702","citing_title":"Seeing Time: Benchmarking Chronological Reasoning and Shortcut Biases in Vision-Language Models","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2606.26734","citing_title":"Robust Onion: Peeling Open Vocab Object Detectors Under Noise","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2511.21025","citing_title":"CaptionQA: Is Your Caption as Useful as the Image Itself?","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2602.04476","citing_title":"Vision-aligned Latent Reasoning for Multi-modal Large Language Model","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11268","citing_title":"Context-Aware Spear Phishing: Generative AI-Enabled Attacks Against Individuals via Public Social Media Data","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06080","citing_title":"MSD-Score: Multi-Scale Distributional Scoring for Reference-Free Image Caption Evaluation","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07394","citing_title":"BalCapRL: A Balanced Framework for RL-Based MLLM Image Captioning","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HGO6IF7RM3RSRSDOVHT7WA5NCQ","json":"https://pith.science/pith/HGO6IF7RM3RSRSDOVHT7WA5NCQ.json","graph_json":"https://pith.science/api/pith-number/HGO6IF7RM3RSRSDOVHT7WA5NCQ/graph.json","events_json":"https://pith.science/api/pith-number/HGO6IF7RM3RSRSDOVHT7WA5NCQ/events.json","paper":"https://pith.science/paper/HGO6IF7R"},"agent_actions":{"view_html":"https://pith.science/pith/HGO6IF7RM3RSRSDOVHT7WA5NCQ","download_json":"https://pith.science/pith/HGO6IF7RM3RSRSDOVHT7WA5NCQ.json","view_paper":"https://pith.science/paper/HGO6IF7R","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.12329&json=true","fetch_graph":"https://pith.science/api/pith-number/HGO6IF7RM3RSRSDOVHT7WA5NCQ/graph.json","fetch_events":"https://pith.science/api/pith-number/HGO6IF7RM3RSRSDOVHT7WA5NCQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HGO6IF7RM3RSRSDOVHT7WA5NCQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HGO6IF7RM3RSRSDOVHT7WA5NCQ/action/storage_attestation","attest_author":"https://pith.science/pith/HGO6IF7RM3RSRSDOVHT7WA5NCQ/action/author_attestation","sign_citation":"https://pith.science/pith/HGO6IF7RM3RSRSDOVHT7WA5NCQ/action/citation_signature","submit_replication":"https://pith.science/pith/HGO6IF7RM3RSRSDOVHT7WA5NCQ/action/replication_record"}},"created_at":"2026-07-05T10:32:10.612882+00:00","updated_at":"2026-07-05T10:32:10.612882+00:00"}