{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:F7DABBN7XLNJQFOS2BDKT5KWKS","short_pith_number":"pith:F7DABBN7","schema_version":"1.0","canonical_sha256":"2fc60085bfbada9815d2d046a9f55654b90b2e97ff2da6c1472975bc570caa48","source":{"kind":"arxiv","id":"2505.20236","version":1},"attestation_state":"computed","paper":{"title":"Seeing is Believing, but How Much? A Comprehensive Analysis of Verbalized Calibration in Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Heli Qi, Junjue Wang, Naoto Yokoya, Qingcheng Zeng, Weihao Xuan","submitted_at":"2025-05-26T17:16:36Z","abstract_excerpt":"Uncertainty quantification is essential for assessing the reliability and trustworthiness of modern AI systems. Among existing approaches, verbalized uncertainty, where models express their confidence through natural language, has emerged as a lightweight and interpretable solution in large language models (LLMs). However, its effectiveness in vision-language models (VLMs) remains insufficiently studied. In this work, we conduct a comprehensive evaluation of verbalized confidence in VLMs, spanning three model categories, four task domains, and three evaluation scenarios. Our results show that "},"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":"2505.20236","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-26T17:16:36Z","cross_cats_sorted":[],"title_canon_sha256":"1911c48591f3f7867d4188ba331e2b5631597c302b936bfb117d704a5518d8ec","abstract_canon_sha256":"96b5a2c3bc3442544c8eeb93d995a591e5d900163c14273623b62934717200e3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:09:54.857541Z","signature_b64":"WYprBLH61gb7pJEFIOX50EIAT1gGurcMzShZYFsd5I/gwMsKq/Gb114yRYeHFp0TDCt6dHpnQc5sI4vN7pjoBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2fc60085bfbada9815d2d046a9f55654b90b2e97ff2da6c1472975bc570caa48","last_reissued_at":"2026-07-05T11:09:54.857034Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:09:54.857034Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Seeing is Believing, but How Much? A Comprehensive Analysis of Verbalized Calibration in Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Heli Qi, Junjue Wang, Naoto Yokoya, Qingcheng Zeng, Weihao Xuan","submitted_at":"2025-05-26T17:16:36Z","abstract_excerpt":"Uncertainty quantification is essential for assessing the reliability and trustworthiness of modern AI systems. Among existing approaches, verbalized uncertainty, where models express their confidence through natural language, has emerged as a lightweight and interpretable solution in large language models (LLMs). However, its effectiveness in vision-language models (VLMs) remains insufficiently studied. In this work, we conduct a comprehensive evaluation of verbalized confidence in VLMs, spanning three model categories, four task domains, and three evaluation scenarios. Our results show that "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.20236","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/2505.20236/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":"2505.20236","created_at":"2026-07-05T11:09:54.857092+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.20236v1","created_at":"2026-07-05T11:09:54.857092+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.20236","created_at":"2026-07-05T11:09:54.857092+00:00"},{"alias_kind":"pith_short_12","alias_value":"F7DABBN7XLNJ","created_at":"2026-07-05T11:09:54.857092+00:00"},{"alias_kind":"pith_short_16","alias_value":"F7DABBN7XLNJQFOS","created_at":"2026-07-05T11:09:54.857092+00:00"},{"alias_kind":"pith_short_8","alias_value":"F7DABBN7","created_at":"2026-07-05T11:09:54.857092+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/F7DABBN7XLNJQFOS2BDKT5KWKS","json":"https://pith.science/pith/F7DABBN7XLNJQFOS2BDKT5KWKS.json","graph_json":"https://pith.science/api/pith-number/F7DABBN7XLNJQFOS2BDKT5KWKS/graph.json","events_json":"https://pith.science/api/pith-number/F7DABBN7XLNJQFOS2BDKT5KWKS/events.json","paper":"https://pith.science/paper/F7DABBN7"},"agent_actions":{"view_html":"https://pith.science/pith/F7DABBN7XLNJQFOS2BDKT5KWKS","download_json":"https://pith.science/pith/F7DABBN7XLNJQFOS2BDKT5KWKS.json","view_paper":"https://pith.science/paper/F7DABBN7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.20236&json=true","fetch_graph":"https://pith.science/api/pith-number/F7DABBN7XLNJQFOS2BDKT5KWKS/graph.json","fetch_events":"https://pith.science/api/pith-number/F7DABBN7XLNJQFOS2BDKT5KWKS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F7DABBN7XLNJQFOS2BDKT5KWKS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F7DABBN7XLNJQFOS2BDKT5KWKS/action/storage_attestation","attest_author":"https://pith.science/pith/F7DABBN7XLNJQFOS2BDKT5KWKS/action/author_attestation","sign_citation":"https://pith.science/pith/F7DABBN7XLNJQFOS2BDKT5KWKS/action/citation_signature","submit_replication":"https://pith.science/pith/F7DABBN7XLNJQFOS2BDKT5KWKS/action/replication_record"}},"created_at":"2026-07-05T11:09:54.857092+00:00","updated_at":"2026-07-05T11:09:54.857092+00:00"}