{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:SPM2RPYVKK2DA4YAGGTP5R6C5I","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":"34be3ce46bee3a225959501c223195bba713b55deb30c6308b71ed974fbe78e6","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-14T15:50:39Z","title_canon_sha256":"597d8df13124af714f792e4b3b78fa6d7516c45a58abb10944944d5e048fdde6"},"schema_version":"1.0","source":{"id":"2504.10342","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.10342","created_at":"2026-07-05T10:56:17Z"},{"alias_kind":"arxiv_version","alias_value":"2504.10342v3","created_at":"2026-07-05T10:56:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.10342","created_at":"2026-07-05T10:56:17Z"},{"alias_kind":"pith_short_12","alias_value":"SPM2RPYVKK2D","created_at":"2026-07-05T10:56:17Z"},{"alias_kind":"pith_short_16","alias_value":"SPM2RPYVKK2DA4YA","created_at":"2026-07-05T10:56:17Z"},{"alias_kind":"pith_short_8","alias_value":"SPM2RPYV","created_at":"2026-07-05T10:56:17Z"}],"graph_snapshots":[{"event_id":"sha256:3e0ed4b396ea60bcf728f687d144fc8f4d9b3152241ca32e2ffd7ea90901d5c9","target":"graph","created_at":"2026-07-05T10:56:17Z","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/2504.10342/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Current multimodal benchmarks often conflate reasoning with domain-specific knowledge, making it difficult to isolate and evaluate general reasoning abilities in non-expert settings. To address this, we introduce VisualPuzzles, a benchmark that targets visual reasoning while deliberately minimizing reliance on specialized knowledge. VisualPuzzles consists of diverse questions spanning five categories: algorithmic, analogical, deductive, inductive, and spatial reasoning. One major source of our questions is manually translated logical reasoning questions from the Chinese Civil Service Examinati","authors_text":"Graham Neubig, Tianyue Ou, Xiang Yue, Yibo Kong, Yueqi Song, Zecheng Li","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-14T15:50:39Z","title":"VisualPuzzles: Decoupling Multimodal Reasoning Evaluation from Domain Knowledge"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.10342","kind":"arxiv","version":3},"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:cc5cbfa09efc8991385ee32481282d58675375266a2372218cd75f51ae28b802","target":"record","created_at":"2026-07-05T10:56:17Z","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":"34be3ce46bee3a225959501c223195bba713b55deb30c6308b71ed974fbe78e6","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-14T15:50:39Z","title_canon_sha256":"597d8df13124af714f792e4b3b78fa6d7516c45a58abb10944944d5e048fdde6"},"schema_version":"1.0","source":{"id":"2504.10342","kind":"arxiv","version":3}},"canonical_sha256":"93d9a8bf1552b430730031a6fec7c2ea1ab1caeb1bd8b2aa8d827939bbfcb750","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"93d9a8bf1552b430730031a6fec7c2ea1ab1caeb1bd8b2aa8d827939bbfcb750","first_computed_at":"2026-07-05T10:56:17.181544Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:56:17.181544Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"47TZPCGWnDX5CwvAwx8t7kn0yvHpgYXWseaxgJsAxFZDNXEhl+lk6j2aTlaDcH9mWDekVEz4Dhsc5t1SX/1wAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:56:17.182039Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.10342","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:cc5cbfa09efc8991385ee32481282d58675375266a2372218cd75f51ae28b802","sha256:3e0ed4b396ea60bcf728f687d144fc8f4d9b3152241ca32e2ffd7ea90901d5c9"],"state_sha256":"ef2fac25cb8d551dbad7b1a5b9e2f14703d470a531f5d7e8350a7654010e374f"}