{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ZDK6QXNGELI6KA3RK4HS3NKDGT","short_pith_number":"pith:ZDK6QXNG","schema_version":"1.0","canonical_sha256":"c8d5e85da622d1e50371570f2db54334d20289a24805cdaeaf563326350696a3","source":{"kind":"arxiv","id":"2608.11074","version":1},"attestation_state":"computed","paper":{"title":"CapProbe: Evaluating Detailed Image Captions via Full-Scene Dense Question Answering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Borui Jiang, Han Shu, Mouxiao Huang, Qiangyu Yan","submitted_at":"2026-08-11T15:36:10Z","abstract_excerpt":"Evaluating detailed image captions from Vision-Language Models (VLMs) requires going beyond surface-level semantic similarity. Reference-based metrics (e.g., CIDEr and SPICE) and LLM-as-scorer protocols struggle to verify dense factual claims, while existing QA-based alternatives generally offer lower probe density, narrower domain coverage, or no explicit alignment between individual questions and segmented image regions. We introduce CapProbe, a full-scene dense QA benchmark that turns detailed caption evaluation into region-aligned factual checking. Each image is decomposed into coarse sema"},"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":"2608.11074","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-08-11T15:36:10Z","cross_cats_sorted":[],"title_canon_sha256":"f6e903e8f84a46fd124e92a42d4627b43f81b3d27dbbf668d5cc1fadc62a0631","abstract_canon_sha256":"56b6457f0762e925d8901f4e6d2c907e8918220d31c5694ad2f5f2c6f59f0ab2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-12T01:24:33.205031Z","signature_b64":"sXrq/PRFA4umSjOKm8sYli31ofiBIqBuYGL6HRdlId7y/qRXJlIl3rGk+esclFYrOuADLFVjtI+PbL65MpHTCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c8d5e85da622d1e50371570f2db54334d20289a24805cdaeaf563326350696a3","last_reissued_at":"2026-08-12T01:24:33.203237Z","signature_status":"signed_v1","first_computed_at":"2026-08-12T01:24:33.203237Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CapProbe: Evaluating Detailed Image Captions via Full-Scene Dense Question Answering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Borui Jiang, Han Shu, Mouxiao Huang, Qiangyu Yan","submitted_at":"2026-08-11T15:36:10Z","abstract_excerpt":"Evaluating detailed image captions from Vision-Language Models (VLMs) requires going beyond surface-level semantic similarity. Reference-based metrics (e.g., CIDEr and SPICE) and LLM-as-scorer protocols struggle to verify dense factual claims, while existing QA-based alternatives generally offer lower probe density, narrower domain coverage, or no explicit alignment between individual questions and segmented image regions. We introduce CapProbe, a full-scene dense QA benchmark that turns detailed caption evaluation into region-aligned factual checking. Each image is decomposed into coarse sema"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.11074","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/2608.11074/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":"2608.11074","created_at":"2026-08-12T01:24:33.207182+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.11074v1","created_at":"2026-08-12T01:24:33.207182+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.11074","created_at":"2026-08-12T01:24:33.207182+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZDK6QXNGELI6","created_at":"2026-08-12T01:24:33.207182+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZDK6QXNGELI6KA3R","created_at":"2026-08-12T01:24:33.207182+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZDK6QXNG","created_at":"2026-08-12T01:24:33.207182+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/ZDK6QXNGELI6KA3RK4HS3NKDGT","json":"https://pith.science/pith/ZDK6QXNGELI6KA3RK4HS3NKDGT.json","graph_json":"https://pith.science/api/pith-number/ZDK6QXNGELI6KA3RK4HS3NKDGT/graph.json","events_json":"https://pith.science/api/pith-number/ZDK6QXNGELI6KA3RK4HS3NKDGT/events.json","paper":"https://pith.science/paper/ZDK6QXNG"},"agent_actions":{"view_html":"https://pith.science/pith/ZDK6QXNGELI6KA3RK4HS3NKDGT","download_json":"https://pith.science/pith/ZDK6QXNGELI6KA3RK4HS3NKDGT.json","view_paper":"https://pith.science/paper/ZDK6QXNG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.11074&json=true","fetch_graph":"https://pith.science/api/pith-number/ZDK6QXNGELI6KA3RK4HS3NKDGT/graph.json","fetch_events":"https://pith.science/api/pith-number/ZDK6QXNGELI6KA3RK4HS3NKDGT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZDK6QXNGELI6KA3RK4HS3NKDGT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZDK6QXNGELI6KA3RK4HS3NKDGT/action/storage_attestation","attest_author":"https://pith.science/pith/ZDK6QXNGELI6KA3RK4HS3NKDGT/action/author_attestation","sign_citation":"https://pith.science/pith/ZDK6QXNGELI6KA3RK4HS3NKDGT/action/citation_signature","submit_replication":"https://pith.science/pith/ZDK6QXNGELI6KA3RK4HS3NKDGT/action/replication_record"}},"created_at":"2026-08-12T01:24:33.207182+00:00","updated_at":"2026-08-12T01:24:33.207182+00:00"}