{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:HHKJUPYOQKWA2DHGBMUDHVFQ5Y","short_pith_number":"pith:HHKJUPYO","canonical_record":{"source":{"id":"2508.20068","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-27T17:22:34Z","cross_cats_sorted":["cs.CV","cs.LG"],"title_canon_sha256":"57be293d80403e5e77bfbfe4aec563235c5d004a547d57749ceeef09d5420af1","abstract_canon_sha256":"d8caf23ebd6d64debfc283ba8a3fe786de95669c9601c47c20c54500f0164df0"},"schema_version":"1.0"},"canonical_sha256":"39d49a3f0e82ac0d0ce60b2833d4b0ee372e417b7e6ea768d4b6dafde82b697a","source":{"kind":"arxiv","id":"2508.20068","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.20068","created_at":"2026-07-05T12:00:23Z"},{"alias_kind":"arxiv_version","alias_value":"2508.20068v1","created_at":"2026-07-05T12:00:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.20068","created_at":"2026-07-05T12:00:23Z"},{"alias_kind":"pith_short_12","alias_value":"HHKJUPYOQKWA","created_at":"2026-07-05T12:00:23Z"},{"alias_kind":"pith_short_16","alias_value":"HHKJUPYOQKWA2DHG","created_at":"2026-07-05T12:00:23Z"},{"alias_kind":"pith_short_8","alias_value":"HHKJUPYO","created_at":"2026-07-05T12:00:23Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:HHKJUPYOQKWA2DHGBMUDHVFQ5Y","target":"record","payload":{"canonical_record":{"source":{"id":"2508.20068","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-27T17:22:34Z","cross_cats_sorted":["cs.CV","cs.LG"],"title_canon_sha256":"57be293d80403e5e77bfbfe4aec563235c5d004a547d57749ceeef09d5420af1","abstract_canon_sha256":"d8caf23ebd6d64debfc283ba8a3fe786de95669c9601c47c20c54500f0164df0"},"schema_version":"1.0"},"canonical_sha256":"39d49a3f0e82ac0d0ce60b2833d4b0ee372e417b7e6ea768d4b6dafde82b697a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:00:23.740702Z","signature_b64":"N8rG44Ep5+oGIsN2dtjfmi3vBJfPBEsv1fxe4GHy7DEvaymhIXrTP4V1F3U7H2HfI4xopYhuikZLUSp4mdhfAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"39d49a3f0e82ac0d0ce60b2833d4b0ee372e417b7e6ea768d4b6dafde82b697a","last_reissued_at":"2026-07-05T12:00:23.740159Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:00:23.740159Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.20068","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":"2026-07-05T12:00:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tI0vfnG5DdQuGOiop1m7PbLuQvGLp5BD42GYaYekt97kEC0zbRnOTy3ACtEVPFbNVoUG0Fv3isF8snLE/GXfAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T17:43:10.910235Z"},"content_sha256":"81a8a554a5e3d0cf33fd9a74b4c88e3fdf5d5c46e412869a61dadba348375dd1","schema_version":"1.0","event_id":"sha256:81a8a554a5e3d0cf33fd9a74b4c88e3fdf5d5c46e412869a61dadba348375dd1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:HHKJUPYOQKWA2DHGBMUDHVFQ5Y","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"11Plus-Bench: Demystifying Multimodal LLM Spatial Reasoning with Cognitive-Inspired Analysis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","cs.LG"],"primary_cat":"cs.CL","authors_text":"Chengzu Li, Furu Wei, Huanyu Zhang, Ivan Vuli\\'c, Jos\\'e Hern\\'andez-Orallo, Qingtao Li, Wenshan Wu, Yan Xia, Zeyu Gao","submitted_at":"2025-08-27T17:22:34Z","abstract_excerpt":"For human cognitive process, spatial reasoning and perception are closely entangled, yet the nature of this interplay remains underexplored in the evaluation of multimodal large language models (MLLMs). While recent MLLM advancements show impressive performance on reasoning, their capacity for human-like spatial cognition remains an open question. In this work, we introduce a systematic evaluation framework to assess the spatial reasoning abilities of state-of-the-art MLLMs relative to human performance. Central to our work is 11Plus-Bench, a high-quality benchmark derived from realistic stand"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.20068","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/2508.20068/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":"2026-07-05T12:00:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oXC11eD2VvK0Z+E0jFQ/gAfdVPU+FVo1Hv8G5ZjctZJkZ5iD/ZXqxzb21TG7tZvV6ZYdxbDlqkI75JCfQTlyDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T17:43:10.911153Z"},"content_sha256":"e3cd79c904cfefb3a4410c961bf19a888ecc2020da2bd58ab2a25577cf398c0f","schema_version":"1.0","event_id":"sha256:e3cd79c904cfefb3a4410c961bf19a888ecc2020da2bd58ab2a25577cf398c0f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HHKJUPYOQKWA2DHGBMUDHVFQ5Y/bundle.json","state_url":"https://pith.science/pith/HHKJUPYOQKWA2DHGBMUDHVFQ5Y/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HHKJUPYOQKWA2DHGBMUDHVFQ5Y/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":"2026-08-15T17:43:10Z","links":{"resolver":"https://pith.science/pith/HHKJUPYOQKWA2DHGBMUDHVFQ5Y","bundle":"https://pith.science/pith/HHKJUPYOQKWA2DHGBMUDHVFQ5Y/bundle.json","state":"https://pith.science/pith/HHKJUPYOQKWA2DHGBMUDHVFQ5Y/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HHKJUPYOQKWA2DHGBMUDHVFQ5Y/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:HHKJUPYOQKWA2DHGBMUDHVFQ5Y","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":"d8caf23ebd6d64debfc283ba8a3fe786de95669c9601c47c20c54500f0164df0","cross_cats_sorted":["cs.CV","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-27T17:22:34Z","title_canon_sha256":"57be293d80403e5e77bfbfe4aec563235c5d004a547d57749ceeef09d5420af1"},"schema_version":"1.0","source":{"id":"2508.20068","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.20068","created_at":"2026-07-05T12:00:23Z"},{"alias_kind":"arxiv_version","alias_value":"2508.20068v1","created_at":"2026-07-05T12:00:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.20068","created_at":"2026-07-05T12:00:23Z"},{"alias_kind":"pith_short_12","alias_value":"HHKJUPYOQKWA","created_at":"2026-07-05T12:00:23Z"},{"alias_kind":"pith_short_16","alias_value":"HHKJUPYOQKWA2DHG","created_at":"2026-07-05T12:00:23Z"},{"alias_kind":"pith_short_8","alias_value":"HHKJUPYO","created_at":"2026-07-05T12:00:23Z"}],"graph_snapshots":[{"event_id":"sha256:e3cd79c904cfefb3a4410c961bf19a888ecc2020da2bd58ab2a25577cf398c0f","target":"graph","created_at":"2026-07-05T12:00:23Z","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/2508.20068/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"For human cognitive process, spatial reasoning and perception are closely entangled, yet the nature of this interplay remains underexplored in the evaluation of multimodal large language models (MLLMs). While recent MLLM advancements show impressive performance on reasoning, their capacity for human-like spatial cognition remains an open question. In this work, we introduce a systematic evaluation framework to assess the spatial reasoning abilities of state-of-the-art MLLMs relative to human performance. Central to our work is 11Plus-Bench, a high-quality benchmark derived from realistic stand","authors_text":"Chengzu Li, Furu Wei, Huanyu Zhang, Ivan Vuli\\'c, Jos\\'e Hern\\'andez-Orallo, Qingtao Li, Wenshan Wu, Yan Xia, Zeyu Gao","cross_cats":["cs.CV","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-27T17:22:34Z","title":"11Plus-Bench: Demystifying Multimodal LLM Spatial Reasoning with Cognitive-Inspired Analysis"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.20068","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":"sha256:81a8a554a5e3d0cf33fd9a74b4c88e3fdf5d5c46e412869a61dadba348375dd1","target":"record","created_at":"2026-07-05T12:00:23Z","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":"d8caf23ebd6d64debfc283ba8a3fe786de95669c9601c47c20c54500f0164df0","cross_cats_sorted":["cs.CV","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-27T17:22:34Z","title_canon_sha256":"57be293d80403e5e77bfbfe4aec563235c5d004a547d57749ceeef09d5420af1"},"schema_version":"1.0","source":{"id":"2508.20068","kind":"arxiv","version":1}},"canonical_sha256":"39d49a3f0e82ac0d0ce60b2833d4b0ee372e417b7e6ea768d4b6dafde82b697a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"39d49a3f0e82ac0d0ce60b2833d4b0ee372e417b7e6ea768d4b6dafde82b697a","first_computed_at":"2026-07-05T12:00:23.740159Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:00:23.740159Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"N8rG44Ep5+oGIsN2dtjfmi3vBJfPBEsv1fxe4GHy7DEvaymhIXrTP4V1F3U7H2HfI4xopYhuikZLUSp4mdhfAA==","signature_status":"signed_v1","signed_at":"2026-07-05T12:00:23.740702Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.20068","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:81a8a554a5e3d0cf33fd9a74b4c88e3fdf5d5c46e412869a61dadba348375dd1","sha256:e3cd79c904cfefb3a4410c961bf19a888ecc2020da2bd58ab2a25577cf398c0f"],"state_sha256":"ff71e97a3681af350340b64b5aa76ee9568115e7ead547feb608ed10bf22c959"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"grViB7VHu5KI3NqOp56QRKn8r0OIUXVGgeGmyowOYk0eIkEPBpn69wGe3ODrlDOQ2f0O4CWbtH/4OKUqSbFQDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-15T17:43:10.917070Z","bundle_sha256":"145bd5f4ec14e362e4773fd6ed74681eb1c8b23c7c1d91c792a0855887888572"}}