{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:SJXFSJWUE7Q765GKELRJADS2MY","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":"1cfac957cd1b6bec9eb6734bd1b1825c8ea7b73efd28a23edc458c4778c742c7","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-07-09T21:44:12Z","title_canon_sha256":"942910af05bc29a821bb20cfd96af2d241f8013eeca9096407d228bc8901adcd"},"schema_version":"1.0","source":{"id":"2507.07297","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.07297","created_at":"2026-07-05T11:34:55Z"},{"alias_kind":"arxiv_version","alias_value":"2507.07297v1","created_at":"2026-07-05T11:34:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.07297","created_at":"2026-07-05T11:34:55Z"},{"alias_kind":"pith_short_12","alias_value":"SJXFSJWUE7Q7","created_at":"2026-07-05T11:34:55Z"},{"alias_kind":"pith_short_16","alias_value":"SJXFSJWUE7Q765GK","created_at":"2026-07-05T11:34:55Z"},{"alias_kind":"pith_short_8","alias_value":"SJXFSJWU","created_at":"2026-07-05T11:34:55Z"}],"graph_snapshots":[{"event_id":"sha256:a7a881b9eed75f01506c6c803a343882eb207b420d84a5c2df8365e17c1f5123","target":"graph","created_at":"2026-07-05T11:34:55Z","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/2507.07297/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advances in large vision-language models have led to impressive performance in visual question answering and multimodal reasoning. However, it remains unclear whether these models genuinely perform grounded visual reasoning or rely on superficial patterns and dataset biases. In this work, we introduce MagiC, a comprehensive benchmark designed to evaluate grounded multimodal cognition, assessing not only answer accuracy but also the quality of step-by-step reasoning and its alignment with relevant visual evidence. Our benchmark includes approximately 5,500 weakly supervised QA examples g","authors_text":"Bingxuan Li, Chengfei Wu, Dan Goldwasser, Fengrong Han, Liqiang Zhang, Ronald Seoh","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-07-09T21:44:12Z","title":"MagiC: Evaluating Multimodal Cognition Toward Grounded Visual Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.07297","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":"sha256:23ae3768dfd3ca4d0e39f671039fa37e07b4c29353d1e848204fec3acd75af8f","target":"record","created_at":"2026-07-05T11:34:55Z","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":"1cfac957cd1b6bec9eb6734bd1b1825c8ea7b73efd28a23edc458c4778c742c7","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-07-09T21:44:12Z","title_canon_sha256":"942910af05bc29a821bb20cfd96af2d241f8013eeca9096407d228bc8901adcd"},"schema_version":"1.0","source":{"id":"2507.07297","kind":"arxiv","version":1}},"canonical_sha256":"926e5926d427e1ff74ca22e2900e5a6608baf4beda87a95e34153f051b6a432d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"926e5926d427e1ff74ca22e2900e5a6608baf4beda87a95e34153f051b6a432d","first_computed_at":"2026-07-05T11:34:55.316520Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:34:55.316520Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"FqfWIfR/xrdcRvmoYagONKKLuDPi/mZedAlzshJ+wgbsLl8gSuzriS4L3c9en1SHWY7Gt90vxBtJClBCVMEgCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:34:55.316929Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.07297","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:23ae3768dfd3ca4d0e39f671039fa37e07b4c29353d1e848204fec3acd75af8f","sha256:a7a881b9eed75f01506c6c803a343882eb207b420d84a5c2df8365e17c1f5123"],"state_sha256":"a396bb7b5e0867503261744bee5365d98e4783fe23ff0e1f1861a72206c191e0"}