{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:B65GY3QCSYWBHEI4RUD447AY3O","short_pith_number":"pith:B65GY3QC","canonical_record":{"source":{"id":"2603.04161","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-03-04T15:19:27Z","cross_cats_sorted":[],"title_canon_sha256":"fdbe916d3c15ec2ef56822b433e9601bb1fb77a548720d7261bf37ab17d26fca","abstract_canon_sha256":"762e97dcf46a716b5ad011eb46e749dd8860448964c96a0c76ef59f582880ad2"},"schema_version":"1.0"},"canonical_sha256":"0fba6c6e02962c13911c8d07ce7c18db99a2eb3f652bbd461ef576b1e7a0a7e8","source":{"kind":"arxiv","id":"2603.04161","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.04161","created_at":"2026-05-20T00:05:42Z"},{"alias_kind":"arxiv_version","alias_value":"2603.04161v2","created_at":"2026-05-20T00:05:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.04161","created_at":"2026-05-20T00:05:42Z"},{"alias_kind":"pith_short_12","alias_value":"B65GY3QCSYWB","created_at":"2026-05-20T00:05:42Z"},{"alias_kind":"pith_short_16","alias_value":"B65GY3QCSYWBHEI4","created_at":"2026-05-20T00:05:42Z"},{"alias_kind":"pith_short_8","alias_value":"B65GY3QC","created_at":"2026-05-20T00:05:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:B65GY3QCSYWBHEI4RUD447AY3O","target":"record","payload":{"canonical_record":{"source":{"id":"2603.04161","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-03-04T15:19:27Z","cross_cats_sorted":[],"title_canon_sha256":"fdbe916d3c15ec2ef56822b433e9601bb1fb77a548720d7261bf37ab17d26fca","abstract_canon_sha256":"762e97dcf46a716b5ad011eb46e749dd8860448964c96a0c76ef59f582880ad2"},"schema_version":"1.0"},"canonical_sha256":"0fba6c6e02962c13911c8d07ce7c18db99a2eb3f652bbd461ef576b1e7a0a7e8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:05:42.753724Z","signature_b64":"Nbmdq1xWRrPG5VYPXj5OJW0S/OQrTxV5/g97OulYMnddgmAbqy4jqg0dREunbfaza4mqplbzexju7I2IqpbmBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0fba6c6e02962c13911c8d07ce7c18db99a2eb3f652bbd461ef576b1e7a0a7e8","last_reissued_at":"2026-05-20T00:05:42.753065Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:05:42.753065Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2603.04161","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:05:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9bGBscwvfzUpCtKgE7kE4v5z9OI6A1fl9SXJSK+au++DVD9cfBo8wy0baCmUBUtSxkLdlR6oylO0BH5fW7Y9Ag==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T14:27:02.793847Z"},"content_sha256":"0e2cdf4e8a6dd68175b9b0fb73e051e13e4a2b6ec2080de4420342a32733ee2c","schema_version":"1.0","event_id":"sha256:0e2cdf4e8a6dd68175b9b0fb73e051e13e4a2b6ec2080de4420342a32733ee2c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:B65GY3QCSYWBHEI4RUD447AY3O","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Traces of Social Competence in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Max van Duijn, Michiel van der Meer, Tom Kouwenhoven","submitted_at":"2026-03-04T15:19:27Z","abstract_excerpt":"The False Belief Test (FBT) has been the main method for assessing Theory of Mind (ToM) and related socio-cognitive competencies. For Large Language Models (LLMs), the reliability and explanatory potential of this test have remained limited due to issues like data contamination, insufficient model details, and inconsistent controls. We address these issues by testing 17 open-weight models on a balanced set of 192 FBT variants (Trott et al., 2023) using Bayesian Logistic regression to identify how model size and post-training affect socio-cognitive competence. We find that scaling model size be"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.04161","kind":"arxiv","version":2},"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/2603.04161/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":"2026-05-20T00:05:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5bnzJZNWy+TprmBTbkOCCNoRXtw2TjHhr9uM2l8RVu2rhCBQ0zUPRMMg1uQSK0IUW2MOzHcHn8+4AKiqtcVmCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T14:27:02.794225Z"},"content_sha256":"b6072d8cae549d31e3e1fb23cbe9784f701ca60e4d12ddc8f2fc136b68508ece","schema_version":"1.0","event_id":"sha256:b6072d8cae549d31e3e1fb23cbe9784f701ca60e4d12ddc8f2fc136b68508ece"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/B65GY3QCSYWBHEI4RUD447AY3O/bundle.json","state_url":"https://pith.science/pith/B65GY3QCSYWBHEI4RUD447AY3O/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/B65GY3QCSYWBHEI4RUD447AY3O/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":"2026-06-11T14:27:02Z","links":{"resolver":"https://pith.science/pith/B65GY3QCSYWBHEI4RUD447AY3O","bundle":"https://pith.science/pith/B65GY3QCSYWBHEI4RUD447AY3O/bundle.json","state":"https://pith.science/pith/B65GY3QCSYWBHEI4RUD447AY3O/state.json","well_known_bundle":"https://pith.science/.well-known/pith/B65GY3QCSYWBHEI4RUD447AY3O/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:B65GY3QCSYWBHEI4RUD447AY3O","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":"762e97dcf46a716b5ad011eb46e749dd8860448964c96a0c76ef59f582880ad2","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-03-04T15:19:27Z","title_canon_sha256":"fdbe916d3c15ec2ef56822b433e9601bb1fb77a548720d7261bf37ab17d26fca"},"schema_version":"1.0","source":{"id":"2603.04161","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.04161","created_at":"2026-05-20T00:05:42Z"},{"alias_kind":"arxiv_version","alias_value":"2603.04161v2","created_at":"2026-05-20T00:05:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.04161","created_at":"2026-05-20T00:05:42Z"},{"alias_kind":"pith_short_12","alias_value":"B65GY3QCSYWB","created_at":"2026-05-20T00:05:42Z"},{"alias_kind":"pith_short_16","alias_value":"B65GY3QCSYWBHEI4","created_at":"2026-05-20T00:05:42Z"},{"alias_kind":"pith_short_8","alias_value":"B65GY3QC","created_at":"2026-05-20T00:05:42Z"}],"graph_snapshots":[{"event_id":"sha256:b6072d8cae549d31e3e1fb23cbe9784f701ca60e4d12ddc8f2fc136b68508ece","target":"graph","created_at":"2026-05-20T00:05:42Z","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/2603.04161/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The False Belief Test (FBT) has been the main method for assessing Theory of Mind (ToM) and related socio-cognitive competencies. For Large Language Models (LLMs), the reliability and explanatory potential of this test have remained limited due to issues like data contamination, insufficient model details, and inconsistent controls. We address these issues by testing 17 open-weight models on a balanced set of 192 FBT variants (Trott et al., 2023) using Bayesian Logistic regression to identify how model size and post-training affect socio-cognitive competence. We find that scaling model size be","authors_text":"Max van Duijn, Michiel van der Meer, Tom Kouwenhoven","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-03-04T15:19:27Z","title":"Traces of Social Competence in Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.04161","kind":"arxiv","version":2},"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":"sha256:0e2cdf4e8a6dd68175b9b0fb73e051e13e4a2b6ec2080de4420342a32733ee2c","target":"record","created_at":"2026-05-20T00:05:42Z","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":"762e97dcf46a716b5ad011eb46e749dd8860448964c96a0c76ef59f582880ad2","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-03-04T15:19:27Z","title_canon_sha256":"fdbe916d3c15ec2ef56822b433e9601bb1fb77a548720d7261bf37ab17d26fca"},"schema_version":"1.0","source":{"id":"2603.04161","kind":"arxiv","version":2}},"canonical_sha256":"0fba6c6e02962c13911c8d07ce7c18db99a2eb3f652bbd461ef576b1e7a0a7e8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0fba6c6e02962c13911c8d07ce7c18db99a2eb3f652bbd461ef576b1e7a0a7e8","first_computed_at":"2026-05-20T00:05:42.753065Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:05:42.753065Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Nbmdq1xWRrPG5VYPXj5OJW0S/OQrTxV5/g97OulYMnddgmAbqy4jqg0dREunbfaza4mqplbzexju7I2IqpbmBw==","signature_status":"signed_v1","signed_at":"2026-05-20T00:05:42.753724Z","signed_message":"canonical_sha256_bytes"},"source_id":"2603.04161","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0e2cdf4e8a6dd68175b9b0fb73e051e13e4a2b6ec2080de4420342a32733ee2c","sha256:b6072d8cae549d31e3e1fb23cbe9784f701ca60e4d12ddc8f2fc136b68508ece"],"state_sha256":"0cbc80913e31242e27de8e93d1b43d7a2027c559d12a8b0686e7a89512fab0df"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FEX1ixUUZ1mvBWu1e/gnXo9EIagWuSatTwzj+P2bN63q5IB1P0dfSy7+KZ+IdKTlL7/yHtiQMFYQPydJnlmOAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-11T14:27:02.796301Z","bundle_sha256":"51477a026ef2b5d807b4593979901cb41922e9ec47a3c6268e51eaf237060d38"}}