{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SPM2RPYVKK2DA4YAGGTP5R6C5I","short_pith_number":"pith:SPM2RPYV","schema_version":"1.0","canonical_sha256":"93d9a8bf1552b430730031a6fec7c2ea1ab1caeb1bd8b2aa8d827939bbfcb750","source":{"kind":"arxiv","id":"2504.10342","version":3},"attestation_state":"computed","paper":{"title":"VisualPuzzles: Decoupling Multimodal Reasoning Evaluation from Domain Knowledge","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Graham Neubig, Tianyue Ou, Xiang Yue, Yibo Kong, Yueqi Song, Zecheng Li","submitted_at":"2025-04-14T15:50:39Z","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"},"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":"2504.10342","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-14T15:50:39Z","cross_cats_sorted":[],"title_canon_sha256":"597d8df13124af714f792e4b3b78fa6d7516c45a58abb10944944d5e048fdde6","abstract_canon_sha256":"34be3ce46bee3a225959501c223195bba713b55deb30c6308b71ed974fbe78e6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:56:17.182039Z","signature_b64":"47TZPCGWnDX5CwvAwx8t7kn0yvHpgYXWseaxgJsAxFZDNXEhl+lk6j2aTlaDcH9mWDekVEz4Dhsc5t1SX/1wAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"93d9a8bf1552b430730031a6fec7c2ea1ab1caeb1bd8b2aa8d827939bbfcb750","last_reissued_at":"2026-07-05T10:56:17.181544Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:56:17.181544Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VisualPuzzles: Decoupling Multimodal Reasoning Evaluation from Domain Knowledge","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Graham Neubig, Tianyue Ou, Xiang Yue, Yibo Kong, Yueqi Song, Zecheng Li","submitted_at":"2025-04-14T15:50:39Z","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"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.10342","kind":"arxiv","version":3},"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/2504.10342/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":"2504.10342","created_at":"2026-07-05T10:56:17.181607+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.10342v3","created_at":"2026-07-05T10:56:17.181607+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.10342","created_at":"2026-07-05T10:56:17.181607+00:00"},{"alias_kind":"pith_short_12","alias_value":"SPM2RPYVKK2D","created_at":"2026-07-05T10:56:17.181607+00:00"},{"alias_kind":"pith_short_16","alias_value":"SPM2RPYVKK2DA4YA","created_at":"2026-07-05T10:56:17.181607+00:00"},{"alias_kind":"pith_short_8","alias_value":"SPM2RPYV","created_at":"2026-07-05T10:56:17.181607+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":13,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22565","citing_title":"Look Light, Think Heavy: What Multimodal Chain-of-Thought Reasoning Can and Cannot Do","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18216","citing_title":"Zone of Proximal Policy Optimization: Teacher in Prompts, Not Gradients","ref_index":141,"is_internal_anchor":false},{"citing_arxiv_id":"2607.02032","citing_title":"PACE: A Proxy for Agentic Capability Evaluation","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19342","citing_title":"Semantic-Enriched Latent Visual Reasoning","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00148","citing_title":"StemBind: When MLLMs Get Lost Between Rules and Instances in Abstract Visual Reasoning","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2602.18600","citing_title":"MapTab: A Diagnostic Benchmark for Long-Horizon Multi-Criteria Multimodal Reasoning on Heterogeneous Topological Graphs","ref_index":65,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19342","citing_title":"Semantic-Enriched Latent Visual Reasoning","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2602.18600","citing_title":"MapTab: A Diagnostic Benchmark for Long-Horizon Multi-Criteria Multimodal Reasoning on Heterogeneous Topological Graphs","ref_index":65,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12163","citing_title":"Self-Consistent Latent Reasoning: Long Latent Sequence Reasoning for Vision-Language Model","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12163","citing_title":"Self-Consistent Latent Reasoning: Long Latent Sequence Reasoning for Vision-Language Model","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09262","citing_title":"Reinforcing Multimodal Reasoning Against Visual Degradation","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10999","citing_title":"TraversalBench: Challenging Paths to Follow for Vision Language Models","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06247","citing_title":"SALLIE: Safeguarding Against Latent Language & Image Exploits","ref_index":25,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SPM2RPYVKK2DA4YAGGTP5R6C5I","json":"https://pith.science/pith/SPM2RPYVKK2DA4YAGGTP5R6C5I.json","graph_json":"https://pith.science/api/pith-number/SPM2RPYVKK2DA4YAGGTP5R6C5I/graph.json","events_json":"https://pith.science/api/pith-number/SPM2RPYVKK2DA4YAGGTP5R6C5I/events.json","paper":"https://pith.science/paper/SPM2RPYV"},"agent_actions":{"view_html":"https://pith.science/pith/SPM2RPYVKK2DA4YAGGTP5R6C5I","download_json":"https://pith.science/pith/SPM2RPYVKK2DA4YAGGTP5R6C5I.json","view_paper":"https://pith.science/paper/SPM2RPYV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.10342&json=true","fetch_graph":"https://pith.science/api/pith-number/SPM2RPYVKK2DA4YAGGTP5R6C5I/graph.json","fetch_events":"https://pith.science/api/pith-number/SPM2RPYVKK2DA4YAGGTP5R6C5I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SPM2RPYVKK2DA4YAGGTP5R6C5I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SPM2RPYVKK2DA4YAGGTP5R6C5I/action/storage_attestation","attest_author":"https://pith.science/pith/SPM2RPYVKK2DA4YAGGTP5R6C5I/action/author_attestation","sign_citation":"https://pith.science/pith/SPM2RPYVKK2DA4YAGGTP5R6C5I/action/citation_signature","submit_replication":"https://pith.science/pith/SPM2RPYVKK2DA4YAGGTP5R6C5I/action/replication_record"}},"created_at":"2026-07-05T10:56:17.181607+00:00","updated_at":"2026-07-05T10:56:17.181607+00:00"}