{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:7CMMHPTEAAIGWGKPC2WM2IYX7I","short_pith_number":"pith:7CMMHPTE","canonical_record":{"source":{"id":"2402.18225","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-28T10:43:54Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"62a90bbdfcbc915e53cc0095c9d2c519e66d65fb75f8052a2f8145989d804ab3","abstract_canon_sha256":"aa25683d98f61507840895bc06ec773d8be2a8de0fc1c16c69c4050af01a7d7b"},"schema_version":"1.0"},"canonical_sha256":"f898c3be6400106b194f16accd2317fa23c6101ffb7513126c6204fd6c2c05d3","source":{"kind":"arxiv","id":"2402.18225","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.18225","created_at":"2026-07-05T07:50:13Z"},{"alias_kind":"arxiv_version","alias_value":"2402.18225v1","created_at":"2026-07-05T07:50:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.18225","created_at":"2026-07-05T07:50:13Z"},{"alias_kind":"pith_short_12","alias_value":"7CMMHPTEAAIG","created_at":"2026-07-05T07:50:13Z"},{"alias_kind":"pith_short_16","alias_value":"7CMMHPTEAAIGWGKP","created_at":"2026-07-05T07:50:13Z"},{"alias_kind":"pith_short_8","alias_value":"7CMMHPTE","created_at":"2026-07-05T07:50:13Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:7CMMHPTEAAIGWGKPC2WM2IYX7I","target":"record","payload":{"canonical_record":{"source":{"id":"2402.18225","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-28T10:43:54Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"62a90bbdfcbc915e53cc0095c9d2c519e66d65fb75f8052a2f8145989d804ab3","abstract_canon_sha256":"aa25683d98f61507840895bc06ec773d8be2a8de0fc1c16c69c4050af01a7d7b"},"schema_version":"1.0"},"canonical_sha256":"f898c3be6400106b194f16accd2317fa23c6101ffb7513126c6204fd6c2c05d3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:50:13.969148Z","signature_b64":"FsyKZmo4vgFZ7cD/BGRCsBhrq2pWyqD5bEZaMaA2SwXP7ev0hkgjw3QGSUCIXFoUv2ZjYAU7sSx4l+WfKAGCCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f898c3be6400106b194f16accd2317fa23c6101ffb7513126c6204fd6c2c05d3","last_reissued_at":"2026-07-05T07:50:13.968718Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:50:13.968718Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.18225","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":"2026-07-05T07:50:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9lw6gQejkZAVzuIgT8paFl7m9/EcxluP4wNNe9fwG1a7JuZFy+xReIKwIgdITpi/OCuZ+9xzbrJ0qKhAvS8sBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T18:59:04.220228Z"},"content_sha256":"627dc0481ec64aed6c346c45da0fbbd93be8bcc83fabc95e28e39b9ff75f2c24","schema_version":"1.0","event_id":"sha256:627dc0481ec64aed6c346c45da0fbbd93be8bcc83fabc95e28e39b9ff75f2c24"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:7CMMHPTEAAIGWGKPC2WM2IYX7I","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CogBench: a large language model walks into a psychology lab","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Eric Schulz, Jane X. Wang, Julian Coda-Forno, Marcel Binz","submitted_at":"2024-02-28T10:43:54Z","abstract_excerpt":"Large language models (LLMs) have significantly advanced the field of artificial intelligence. Yet, evaluating them comprehensively remains challenging. We argue that this is partly due to the predominant focus on performance metrics in most benchmarks. This paper introduces CogBench, a benchmark that includes ten behavioral metrics derived from seven cognitive psychology experiments. This novel approach offers a toolkit for phenotyping LLMs' behavior. We apply CogBench to 35 LLMs, yielding a rich and diverse dataset. We analyze this data using statistical multilevel modeling techniques, accou"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.18225","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/2402.18225/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":"2026-07-05T07:50:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yiHt0bLluBC80ttBznVQ8cYnDFHbLFytzrC9ias8OSukNG9UkUfKX7HpP/VoS4hpjxyCqb/4lTTprJr1oYyxCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T18:59:04.220750Z"},"content_sha256":"ed63082f61d4e1315131270d451e28aa4aeb246fa63e10066b7f82134a0e4334","schema_version":"1.0","event_id":"sha256:ed63082f61d4e1315131270d451e28aa4aeb246fa63e10066b7f82134a0e4334"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7CMMHPTEAAIGWGKPC2WM2IYX7I/bundle.json","state_url":"https://pith.science/pith/7CMMHPTEAAIGWGKPC2WM2IYX7I/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7CMMHPTEAAIGWGKPC2WM2IYX7I/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":"2026-08-08T18:59:04Z","links":{"resolver":"https://pith.science/pith/7CMMHPTEAAIGWGKPC2WM2IYX7I","bundle":"https://pith.science/pith/7CMMHPTEAAIGWGKPC2WM2IYX7I/bundle.json","state":"https://pith.science/pith/7CMMHPTEAAIGWGKPC2WM2IYX7I/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7CMMHPTEAAIGWGKPC2WM2IYX7I/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:7CMMHPTEAAIGWGKPC2WM2IYX7I","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":"aa25683d98f61507840895bc06ec773d8be2a8de0fc1c16c69c4050af01a7d7b","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-28T10:43:54Z","title_canon_sha256":"62a90bbdfcbc915e53cc0095c9d2c519e66d65fb75f8052a2f8145989d804ab3"},"schema_version":"1.0","source":{"id":"2402.18225","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.18225","created_at":"2026-07-05T07:50:13Z"},{"alias_kind":"arxiv_version","alias_value":"2402.18225v1","created_at":"2026-07-05T07:50:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.18225","created_at":"2026-07-05T07:50:13Z"},{"alias_kind":"pith_short_12","alias_value":"7CMMHPTEAAIG","created_at":"2026-07-05T07:50:13Z"},{"alias_kind":"pith_short_16","alias_value":"7CMMHPTEAAIGWGKP","created_at":"2026-07-05T07:50:13Z"},{"alias_kind":"pith_short_8","alias_value":"7CMMHPTE","created_at":"2026-07-05T07:50:13Z"}],"graph_snapshots":[{"event_id":"sha256:ed63082f61d4e1315131270d451e28aa4aeb246fa63e10066b7f82134a0e4334","target":"graph","created_at":"2026-07-05T07:50:13Z","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/2402.18225/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have significantly advanced the field of artificial intelligence. Yet, evaluating them comprehensively remains challenging. We argue that this is partly due to the predominant focus on performance metrics in most benchmarks. This paper introduces CogBench, a benchmark that includes ten behavioral metrics derived from seven cognitive psychology experiments. This novel approach offers a toolkit for phenotyping LLMs' behavior. We apply CogBench to 35 LLMs, yielding a rich and diverse dataset. We analyze this data using statistical multilevel modeling techniques, accou","authors_text":"Eric Schulz, Jane X. Wang, Julian Coda-Forno, Marcel Binz","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-28T10:43:54Z","title":"CogBench: a large language model walks into a psychology lab"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.18225","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":"sha256:627dc0481ec64aed6c346c45da0fbbd93be8bcc83fabc95e28e39b9ff75f2c24","target":"record","created_at":"2026-07-05T07:50:13Z","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":"aa25683d98f61507840895bc06ec773d8be2a8de0fc1c16c69c4050af01a7d7b","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-28T10:43:54Z","title_canon_sha256":"62a90bbdfcbc915e53cc0095c9d2c519e66d65fb75f8052a2f8145989d804ab3"},"schema_version":"1.0","source":{"id":"2402.18225","kind":"arxiv","version":1}},"canonical_sha256":"f898c3be6400106b194f16accd2317fa23c6101ffb7513126c6204fd6c2c05d3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f898c3be6400106b194f16accd2317fa23c6101ffb7513126c6204fd6c2c05d3","first_computed_at":"2026-07-05T07:50:13.968718Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:50:13.968718Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"FsyKZmo4vgFZ7cD/BGRCsBhrq2pWyqD5bEZaMaA2SwXP7ev0hkgjw3QGSUCIXFoUv2ZjYAU7sSx4l+WfKAGCCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:50:13.969148Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.18225","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:627dc0481ec64aed6c346c45da0fbbd93be8bcc83fabc95e28e39b9ff75f2c24","sha256:ed63082f61d4e1315131270d451e28aa4aeb246fa63e10066b7f82134a0e4334"],"state_sha256":"0a558f2517d0ba328055dea3b3b86059deaa0d734e4ca921b225303d0a497b76"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8oNRC/ILX3U4chvzRaBZlyys+3RQ7/u7EBO2/GSMVTd2lWNrubpnHbXl9Hy2Y6CrB9+X5Z1NYigIlb/Fuu82Dw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T18:59:04.225390Z","bundle_sha256":"2477a55218e0382f3faef3a224957e96144fb9286cb80b9e9f84aabdc8f06c56"}}