{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:SYENZCEB2ZJY6MICHLHLYT7UYR","short_pith_number":"pith:SYENZCEB","canonical_record":{"source":{"id":"2402.18409","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-02-28T15:28:36Z","cross_cats_sorted":["cs.CL","cs.CV"],"title_canon_sha256":"99045731dd51a9aae5226bc392945c46ec233b891b2c7ceac2d816723692d74a","abstract_canon_sha256":"79ca3b739fef153637b6749a28220c933e799d743dab8ebebf334812f1cdabdc"},"schema_version":"1.0"},"canonical_sha256":"9608dc8881d6538f31023acebc4ff4c475a67b3cbfaa9b295e6001ea0eb06dd3","source":{"kind":"arxiv","id":"2402.18409","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.18409","created_at":"2026-07-05T10:13:38Z"},{"alias_kind":"arxiv_version","alias_value":"2402.18409v4","created_at":"2026-07-05T10:13:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.18409","created_at":"2026-07-05T10:13:38Z"},{"alias_kind":"pith_short_12","alias_value":"SYENZCEB2ZJY","created_at":"2026-07-05T10:13:38Z"},{"alias_kind":"pith_short_16","alias_value":"SYENZCEB2ZJY6MIC","created_at":"2026-07-05T10:13:38Z"},{"alias_kind":"pith_short_8","alias_value":"SYENZCEB","created_at":"2026-07-05T10:13:38Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:SYENZCEB2ZJY6MICHLHLYT7UYR","target":"record","payload":{"canonical_record":{"source":{"id":"2402.18409","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-02-28T15:28:36Z","cross_cats_sorted":["cs.CL","cs.CV"],"title_canon_sha256":"99045731dd51a9aae5226bc392945c46ec233b891b2c7ceac2d816723692d74a","abstract_canon_sha256":"79ca3b739fef153637b6749a28220c933e799d743dab8ebebf334812f1cdabdc"},"schema_version":"1.0"},"canonical_sha256":"9608dc8881d6538f31023acebc4ff4c475a67b3cbfaa9b295e6001ea0eb06dd3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:13:38.935934Z","signature_b64":"6vpR+j7pLAvYnEP5d1jfRAcdBWOI0Fw9Hir2o5kG00sG+6zSHwHFc43uzLwlCxIts00w3KOGIFp+ru6Uoy/PDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9608dc8881d6538f31023acebc4ff4c475a67b3cbfaa9b295e6001ea0eb06dd3","last_reissued_at":"2026-07-05T10:13:38.935446Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:13:38.935446Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.18409","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:13:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JuT+xxbdzNKRHte9BrsyuJLM+VhTXLXVPibaD1sTiUHRdl071inbbwyCCiDPXfsWpXw3YNVHMjIGjBJ0Kf6TDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T11:17:07.952473Z"},"content_sha256":"b13997ebef0d732f63b61d0e452efb6cc46b93892388925d5c83d1831dce7b6e","schema_version":"1.0","event_id":"sha256:b13997ebef0d732f63b61d0e452efb6cc46b93892388925d5c83d1831dce7b6e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:SYENZCEB2ZJY6MICHLHLYT7UYR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A Cognitive Evaluation Benchmark of Image Reasoning and Description for Large Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.AI","authors_text":"Chunhao Zhang, Kenny Q. Zhu, Mengyue Wu, Xiujie Song, Yanyi Chen","submitted_at":"2024-02-28T15:28:36Z","abstract_excerpt":"Large Vision-Language Models (LVLMs), despite their recent success, are hardly comprehensively tested for their cognitive abilities. Inspired by the prevalent use of the Cookie Theft task in human cognitive tests, we propose a novel evaluation benchmark to evaluate high-level cognitive abilities of LVLMs using images with rich semantics. The benchmark consists of 251 images along with comprehensive annotations. It defines eight reasoning capabilities and comprises an image description task and a visual question answering task. Our evaluation of well-known LVLMs shows that there is still a sign"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.18409","kind":"arxiv","version":4},"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/2402.18409/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":"2026-07-05T10:13:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gJbsib3JFh/eKEdOiAXqTB/h6DCIeoLf3iYANhoD0a5dDqa6+OnCFOuJC3nOiUXXEND/Pmbna8cpiPJ1s8uvCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T11:17:07.952982Z"},"content_sha256":"2e8d35c70aeebf2abd84a2013d46c483a0f5c611ac4e6eab877227f42c70fbb6","schema_version":"1.0","event_id":"sha256:2e8d35c70aeebf2abd84a2013d46c483a0f5c611ac4e6eab877227f42c70fbb6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SYENZCEB2ZJY6MICHLHLYT7UYR/bundle.json","state_url":"https://pith.science/pith/SYENZCEB2ZJY6MICHLHLYT7UYR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SYENZCEB2ZJY6MICHLHLYT7UYR/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":"2026-08-04T11:17:07Z","links":{"resolver":"https://pith.science/pith/SYENZCEB2ZJY6MICHLHLYT7UYR","bundle":"https://pith.science/pith/SYENZCEB2ZJY6MICHLHLYT7UYR/bundle.json","state":"https://pith.science/pith/SYENZCEB2ZJY6MICHLHLYT7UYR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SYENZCEB2ZJY6MICHLHLYT7UYR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:SYENZCEB2ZJY6MICHLHLYT7UYR","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":"79ca3b739fef153637b6749a28220c933e799d743dab8ebebf334812f1cdabdc","cross_cats_sorted":["cs.CL","cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-02-28T15:28:36Z","title_canon_sha256":"99045731dd51a9aae5226bc392945c46ec233b891b2c7ceac2d816723692d74a"},"schema_version":"1.0","source":{"id":"2402.18409","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.18409","created_at":"2026-07-05T10:13:38Z"},{"alias_kind":"arxiv_version","alias_value":"2402.18409v4","created_at":"2026-07-05T10:13:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.18409","created_at":"2026-07-05T10:13:38Z"},{"alias_kind":"pith_short_12","alias_value":"SYENZCEB2ZJY","created_at":"2026-07-05T10:13:38Z"},{"alias_kind":"pith_short_16","alias_value":"SYENZCEB2ZJY6MIC","created_at":"2026-07-05T10:13:38Z"},{"alias_kind":"pith_short_8","alias_value":"SYENZCEB","created_at":"2026-07-05T10:13:38Z"}],"graph_snapshots":[{"event_id":"sha256:2e8d35c70aeebf2abd84a2013d46c483a0f5c611ac4e6eab877227f42c70fbb6","target":"graph","created_at":"2026-07-05T10:13:38Z","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/2402.18409/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Vision-Language Models (LVLMs), despite their recent success, are hardly comprehensively tested for their cognitive abilities. Inspired by the prevalent use of the Cookie Theft task in human cognitive tests, we propose a novel evaluation benchmark to evaluate high-level cognitive abilities of LVLMs using images with rich semantics. The benchmark consists of 251 images along with comprehensive annotations. It defines eight reasoning capabilities and comprises an image description task and a visual question answering task. Our evaluation of well-known LVLMs shows that there is still a sign","authors_text":"Chunhao Zhang, Kenny Q. Zhu, Mengyue Wu, Xiujie Song, Yanyi Chen","cross_cats":["cs.CL","cs.CV"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-02-28T15:28:36Z","title":"A Cognitive Evaluation Benchmark of Image Reasoning and Description for Large Vision-Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.18409","kind":"arxiv","version":4},"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:b13997ebef0d732f63b61d0e452efb6cc46b93892388925d5c83d1831dce7b6e","target":"record","created_at":"2026-07-05T10:13:38Z","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":"79ca3b739fef153637b6749a28220c933e799d743dab8ebebf334812f1cdabdc","cross_cats_sorted":["cs.CL","cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-02-28T15:28:36Z","title_canon_sha256":"99045731dd51a9aae5226bc392945c46ec233b891b2c7ceac2d816723692d74a"},"schema_version":"1.0","source":{"id":"2402.18409","kind":"arxiv","version":4}},"canonical_sha256":"9608dc8881d6538f31023acebc4ff4c475a67b3cbfaa9b295e6001ea0eb06dd3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9608dc8881d6538f31023acebc4ff4c475a67b3cbfaa9b295e6001ea0eb06dd3","first_computed_at":"2026-07-05T10:13:38.935446Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:13:38.935446Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6vpR+j7pLAvYnEP5d1jfRAcdBWOI0Fw9Hir2o5kG00sG+6zSHwHFc43uzLwlCxIts00w3KOGIFp+ru6Uoy/PDg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:13:38.935934Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.18409","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b13997ebef0d732f63b61d0e452efb6cc46b93892388925d5c83d1831dce7b6e","sha256:2e8d35c70aeebf2abd84a2013d46c483a0f5c611ac4e6eab877227f42c70fbb6"],"state_sha256":"d5d3ec808b44937d1b2adee88e72a144c50f6a4f61651d48eb61edcaf3242d4a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vaE7NBTqz7vdVQPEg1rhiUSLLdOngakjmC2OPNcGk6qhHgJpcgPVEe5qUuBhVrSrl5E9FkdRZeCcs2zTsIunDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T11:17:07.957353Z","bundle_sha256":"c04e154c50ddc3f900bafa8dca6b46ca53cd3a7c99ba5f722b7b1763a0ebdfad"}}