{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:2NDS3QTKMQMJMBZOH7EW366F3G","short_pith_number":"pith:2NDS3QTK","canonical_record":{"source":{"id":"2605.28023","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-27T06:27:04Z","cross_cats_sorted":["cs.AI","cs.CL","cs.MM"],"title_canon_sha256":"43e93e1ec97d573d56785b75df91547705a5408d85317444472072574595ca31","abstract_canon_sha256":"662a9ed723b1be9332368e4e58b7be0ceb1ab1b120bc19cf2ebd0a01204c3082"},"schema_version":"1.0"},"canonical_sha256":"d3472dc26a641896072e3fc96dfbc5d9be0d878e20dff93be0f1027c6cdb2395","source":{"kind":"arxiv","id":"2605.28023","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.28023","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"arxiv_version","alias_value":"2605.28023v1","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.28023","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"pith_short_12","alias_value":"2NDS3QTKMQMJ","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"pith_short_16","alias_value":"2NDS3QTKMQMJMBZO","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"pith_short_8","alias_value":"2NDS3QTK","created_at":"2026-05-28T01:04:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:2NDS3QTKMQMJMBZOH7EW366F3G","target":"record","payload":{"canonical_record":{"source":{"id":"2605.28023","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-27T06:27:04Z","cross_cats_sorted":["cs.AI","cs.CL","cs.MM"],"title_canon_sha256":"43e93e1ec97d573d56785b75df91547705a5408d85317444472072574595ca31","abstract_canon_sha256":"662a9ed723b1be9332368e4e58b7be0ceb1ab1b120bc19cf2ebd0a01204c3082"},"schema_version":"1.0"},"canonical_sha256":"d3472dc26a641896072e3fc96dfbc5d9be0d878e20dff93be0f1027c6cdb2395","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:04:56.319826Z","signature_b64":"O79O6awZk9VlTLLxx63LLViC9QWzDzFP8gYsOxjXlKbKw8w8oKOHameLpFYb0AN03TxdZgrOfP5lhCMJJOfEDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d3472dc26a641896072e3fc96dfbc5d9be0d878e20dff93be0f1027c6cdb2395","last_reissued_at":"2026-05-28T01:04:56.319411Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:04:56.319411Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.28023","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-28T01:04:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pLTLid7HywhkblMeG7rAWra5BJ7JnAQ1+f1SLBHzN47ZwAnpDHQavG62c+BV9mBN4SM84Hg0LGOKd+poNpnCCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T19:13:49.431769Z"},"content_sha256":"43f30cba01c7adb7ae122dc0df58f6f3c07b2435160c645ad2ff18ca512fade5","schema_version":"1.0","event_id":"sha256:43f30cba01c7adb7ae122dc0df58f6f3c07b2435160c645ad2ff18ca512fade5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:2NDS3QTKMQMJMBZOH7EW366F3G","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"VCap: Hypergeometric Rewards for Weak-to-Strong Visual Captioning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.MM"],"primary_cat":"cs.CV","authors_text":"Bin Wen, Changyi Liu, Chun Yuan, Fan Yang, Han Li, Haonan Fan, Jinpeng Wang, Kaiyu Jiang, Tianke Zhang, Tingting Gao, Xingyu Lu, Xuanyu Zheng, Yancheng Long, Yankai Yang, Yi-Fan Zhang, Yiyang Fan","submitted_at":"2026-05-27T06:27:04Z","abstract_excerpt":"Visual captioning requires models to capture visual content faithfully while minimizing both omission and hallucination. As the dominant paradigm for captioning, MLLMs have achieved strong performance through scaling and high-quality data. Recently, RL has emerged as a key route to driving MLLMs toward higher precision and broader coverage, however, existing reward designs for captioning fail to provide fine-grained and reliable signals for factual verification, limiting their effectiveness. To address this, we propose VCap, a Witness-Adjudicator reward that pairs the reference caption (a witn"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.28023","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/2605.28023/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-28T01:04:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EK44Fz1J9HxBZ57q0J6f3CyzIOhzy+4IvGzStfqwd6a0j9Ka1eXS3xHIA5404O7pNY+O2YhEx1CPTFhHSGUoBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T19:13:49.432174Z"},"content_sha256":"26301818659684b4da0b231cd58f519d71f21a9be805d3f8c9bf54ece2ef6027","schema_version":"1.0","event_id":"sha256:26301818659684b4da0b231cd58f519d71f21a9be805d3f8c9bf54ece2ef6027"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/2NDS3QTKMQMJMBZOH7EW366F3G/bundle.json","state_url":"https://pith.science/pith/2NDS3QTKMQMJMBZOH7EW366F3G/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/2NDS3QTKMQMJMBZOH7EW366F3G/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-28T19:13:49Z","links":{"resolver":"https://pith.science/pith/2NDS3QTKMQMJMBZOH7EW366F3G","bundle":"https://pith.science/pith/2NDS3QTKMQMJMBZOH7EW366F3G/bundle.json","state":"https://pith.science/pith/2NDS3QTKMQMJMBZOH7EW366F3G/state.json","well_known_bundle":"https://pith.science/.well-known/pith/2NDS3QTKMQMJMBZOH7EW366F3G/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:2NDS3QTKMQMJMBZOH7EW366F3G","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"662a9ed723b1be9332368e4e58b7be0ceb1ab1b120bc19cf2ebd0a01204c3082","cross_cats_sorted":["cs.AI","cs.CL","cs.MM"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-27T06:27:04Z","title_canon_sha256":"43e93e1ec97d573d56785b75df91547705a5408d85317444472072574595ca31"},"schema_version":"1.0","source":{"id":"2605.28023","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.28023","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"arxiv_version","alias_value":"2605.28023v1","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.28023","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"pith_short_12","alias_value":"2NDS3QTKMQMJ","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"pith_short_16","alias_value":"2NDS3QTKMQMJMBZO","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"pith_short_8","alias_value":"2NDS3QTK","created_at":"2026-05-28T01:04:56Z"}],"graph_snapshots":[{"event_id":"sha256:26301818659684b4da0b231cd58f519d71f21a9be805d3f8c9bf54ece2ef6027","target":"graph","created_at":"2026-05-28T01:04:56Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2605.28023/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Visual captioning requires models to capture visual content faithfully while minimizing both omission and hallucination. As the dominant paradigm for captioning, MLLMs have achieved strong performance through scaling and high-quality data. Recently, RL has emerged as a key route to driving MLLMs toward higher precision and broader coverage, however, existing reward designs for captioning fail to provide fine-grained and reliable signals for factual verification, limiting their effectiveness. To address this, we propose VCap, a Witness-Adjudicator reward that pairs the reference caption (a witn","authors_text":"Bin Wen, Changyi Liu, Chun Yuan, Fan Yang, Han Li, Haonan Fan, Jinpeng Wang, Kaiyu Jiang, Tianke Zhang, Tingting Gao, Xingyu Lu, Xuanyu Zheng, Yancheng Long, Yankai Yang, Yi-Fan Zhang, Yiyang Fan","cross_cats":["cs.AI","cs.CL","cs.MM"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-27T06:27:04Z","title":"VCap: Hypergeometric Rewards for Weak-to-Strong Visual Captioning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.28023","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:43f30cba01c7adb7ae122dc0df58f6f3c07b2435160c645ad2ff18ca512fade5","target":"record","created_at":"2026-05-28T01:04:56Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"662a9ed723b1be9332368e4e58b7be0ceb1ab1b120bc19cf2ebd0a01204c3082","cross_cats_sorted":["cs.AI","cs.CL","cs.MM"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-27T06:27:04Z","title_canon_sha256":"43e93e1ec97d573d56785b75df91547705a5408d85317444472072574595ca31"},"schema_version":"1.0","source":{"id":"2605.28023","kind":"arxiv","version":1}},"canonical_sha256":"d3472dc26a641896072e3fc96dfbc5d9be0d878e20dff93be0f1027c6cdb2395","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d3472dc26a641896072e3fc96dfbc5d9be0d878e20dff93be0f1027c6cdb2395","first_computed_at":"2026-05-28T01:04:56.319411Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-28T01:04:56.319411Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"O79O6awZk9VlTLLxx63LLViC9QWzDzFP8gYsOxjXlKbKw8w8oKOHameLpFYb0AN03TxdZgrOfP5lhCMJJOfEDQ==","signature_status":"signed_v1","signed_at":"2026-05-28T01:04:56.319826Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.28023","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:43f30cba01c7adb7ae122dc0df58f6f3c07b2435160c645ad2ff18ca512fade5","sha256:26301818659684b4da0b231cd58f519d71f21a9be805d3f8c9bf54ece2ef6027"],"state_sha256":"882eb6d04a0bb916ef48fb246964d15db70b934a67efceea17556b3daac745fe"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FlTKOfbO1QwcoAh62YD8Xit9B4nK8TaMgJ8RnwZcBws0rdmn0QAMiR5qB1w1PUK1J0AtvQzEXwiAbQxxr2hHCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-28T19:13:49.434331Z","bundle_sha256":"8e48f04b63def7808c7cb4597ad4644dcfbcec7c212bdea6ad4c4a32e2ebd190"}}