{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:OJUVE6OCHLDVDI4ZY2MSUMV3YP","short_pith_number":"pith:OJUVE6OC","canonical_record":{"source":{"id":"2309.05452","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-11T13:47:07Z","cross_cats_sorted":[],"title_canon_sha256":"cc48b2268bd953d1a571e4658688a5b22a2c19c8f266c4c641fe9f7dfd7bcfd2","abstract_canon_sha256":"3f06333acc5af32b5b7f3aebd8344d83a723ce3530e3f22b7a42becf90447c36"},"schema_version":"1.0"},"canonical_sha256":"72695279c23ac751a399c6992a32bbc3fbf6bd15dc36975000262a15b71b0125","source":{"kind":"arxiv","id":"2309.05452","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.05452","created_at":"2026-07-05T08:07:58Z"},{"alias_kind":"arxiv_version","alias_value":"2309.05452v2","created_at":"2026-07-05T08:07:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.05452","created_at":"2026-07-05T08:07:58Z"},{"alias_kind":"pith_short_12","alias_value":"OJUVE6OCHLDV","created_at":"2026-07-05T08:07:58Z"},{"alias_kind":"pith_short_16","alias_value":"OJUVE6OCHLDVDI4Z","created_at":"2026-07-05T08:07:58Z"},{"alias_kind":"pith_short_8","alias_value":"OJUVE6OC","created_at":"2026-07-05T08:07:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:OJUVE6OCHLDVDI4ZY2MSUMV3YP","target":"record","payload":{"canonical_record":{"source":{"id":"2309.05452","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-11T13:47:07Z","cross_cats_sorted":[],"title_canon_sha256":"cc48b2268bd953d1a571e4658688a5b22a2c19c8f266c4c641fe9f7dfd7bcfd2","abstract_canon_sha256":"3f06333acc5af32b5b7f3aebd8344d83a723ce3530e3f22b7a42becf90447c36"},"schema_version":"1.0"},"canonical_sha256":"72695279c23ac751a399c6992a32bbc3fbf6bd15dc36975000262a15b71b0125","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:07:58.072716Z","signature_b64":"XkfWdmlsrpiyM5gdR7KaBvkM/jxGQBfokj7Z6KykXWc74puHmsvO/oRNuEehaNBQVAlW8A7OuXmMAIocDGYvBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"72695279c23ac751a399c6992a32bbc3fbf6bd15dc36975000262a15b71b0125","last_reissued_at":"2026-07-05T08:07:58.072302Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:07:58.072302Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2309.05452","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":"2026-07-05T08:07:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AcyIROgMcaJ3FE8cCUexM/XbIFl1Xl144FuHak91UQBZ1viu2BvpbaTuh6ufpcftg894WHnR/x8Rg0RIBWE1Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T18:31:11.678562Z"},"content_sha256":"d9c0292edc1dc491755da46d843c2c2ebabdcccc86364163d9b0ca45b41e87c7","schema_version":"1.0","event_id":"sha256:d9c0292edc1dc491755da46d843c2c2ebabdcccc86364163d9b0ca45b41e87c7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:OJUVE6OCHLDVDI4ZY2MSUMV3YP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Evaluating the Deductive Competence of Large Language Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Spencer M. Seals, Valerie L. Shalin","submitted_at":"2023-09-11T13:47:07Z","abstract_excerpt":"The development of highly fluent large language models (LLMs) has prompted increased interest in assessing their reasoning and problem-solving capabilities. We investigate whether several LLMs can solve a classic type of deductive reasoning problem from the cognitive science literature. The tested LLMs have limited abilities to solve these problems in their conventional form. We performed follow up experiments to investigate if changes to the presentation format and content improve model performance. We do find performance differences between conditions; however, they do not improve overall pe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.05452","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/2309.05452/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":"2026-07-05T08:07:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"u06uG96K9eN5jVmKQw6kR3mX9uPl/V6QaKedERuzwMLQh15XPMEHzeP4ydfa/UhY2fCJfemnAKlpkdMOkhvgCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T18:31:11.679060Z"},"content_sha256":"c8729d23880410e5b90ae2d18b9f333944e3ec3a287c68b1eaa55c4ceb346ee7","schema_version":"1.0","event_id":"sha256:c8729d23880410e5b90ae2d18b9f333944e3ec3a287c68b1eaa55c4ceb346ee7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OJUVE6OCHLDVDI4ZY2MSUMV3YP/bundle.json","state_url":"https://pith.science/pith/OJUVE6OCHLDVDI4ZY2MSUMV3YP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OJUVE6OCHLDVDI4ZY2MSUMV3YP/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":"2026-08-16T18:31:11Z","links":{"resolver":"https://pith.science/pith/OJUVE6OCHLDVDI4ZY2MSUMV3YP","bundle":"https://pith.science/pith/OJUVE6OCHLDVDI4ZY2MSUMV3YP/bundle.json","state":"https://pith.science/pith/OJUVE6OCHLDVDI4ZY2MSUMV3YP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OJUVE6OCHLDVDI4ZY2MSUMV3YP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:OJUVE6OCHLDVDI4ZY2MSUMV3YP","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":"3f06333acc5af32b5b7f3aebd8344d83a723ce3530e3f22b7a42becf90447c36","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-11T13:47:07Z","title_canon_sha256":"cc48b2268bd953d1a571e4658688a5b22a2c19c8f266c4c641fe9f7dfd7bcfd2"},"schema_version":"1.0","source":{"id":"2309.05452","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.05452","created_at":"2026-07-05T08:07:58Z"},{"alias_kind":"arxiv_version","alias_value":"2309.05452v2","created_at":"2026-07-05T08:07:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.05452","created_at":"2026-07-05T08:07:58Z"},{"alias_kind":"pith_short_12","alias_value":"OJUVE6OCHLDV","created_at":"2026-07-05T08:07:58Z"},{"alias_kind":"pith_short_16","alias_value":"OJUVE6OCHLDVDI4Z","created_at":"2026-07-05T08:07:58Z"},{"alias_kind":"pith_short_8","alias_value":"OJUVE6OC","created_at":"2026-07-05T08:07:58Z"}],"graph_snapshots":[{"event_id":"sha256:c8729d23880410e5b90ae2d18b9f333944e3ec3a287c68b1eaa55c4ceb346ee7","target":"graph","created_at":"2026-07-05T08:07:58Z","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/2309.05452/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The development of highly fluent large language models (LLMs) has prompted increased interest in assessing their reasoning and problem-solving capabilities. We investigate whether several LLMs can solve a classic type of deductive reasoning problem from the cognitive science literature. The tested LLMs have limited abilities to solve these problems in their conventional form. We performed follow up experiments to investigate if changes to the presentation format and content improve model performance. We do find performance differences between conditions; however, they do not improve overall pe","authors_text":"Spencer M. Seals, Valerie L. Shalin","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-11T13:47:07Z","title":"Evaluating the Deductive Competence of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.05452","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:d9c0292edc1dc491755da46d843c2c2ebabdcccc86364163d9b0ca45b41e87c7","target":"record","created_at":"2026-07-05T08:07:58Z","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":"3f06333acc5af32b5b7f3aebd8344d83a723ce3530e3f22b7a42becf90447c36","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-11T13:47:07Z","title_canon_sha256":"cc48b2268bd953d1a571e4658688a5b22a2c19c8f266c4c641fe9f7dfd7bcfd2"},"schema_version":"1.0","source":{"id":"2309.05452","kind":"arxiv","version":2}},"canonical_sha256":"72695279c23ac751a399c6992a32bbc3fbf6bd15dc36975000262a15b71b0125","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"72695279c23ac751a399c6992a32bbc3fbf6bd15dc36975000262a15b71b0125","first_computed_at":"2026-07-05T08:07:58.072302Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:07:58.072302Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XkfWdmlsrpiyM5gdR7KaBvkM/jxGQBfokj7Z6KykXWc74puHmsvO/oRNuEehaNBQVAlW8A7OuXmMAIocDGYvBw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:07:58.072716Z","signed_message":"canonical_sha256_bytes"},"source_id":"2309.05452","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d9c0292edc1dc491755da46d843c2c2ebabdcccc86364163d9b0ca45b41e87c7","sha256:c8729d23880410e5b90ae2d18b9f333944e3ec3a287c68b1eaa55c4ceb346ee7"],"state_sha256":"a34660ff2226b8979bed76e4868b9047a10b596162874fa121a1f1de69963b4a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VWIPyjsagOoVycGFWEfHPr48djk5jQ00MXY+uHYDnvM9m+P786DZ+5NRtTM48+4KqK7rQcDlAOqEVmFFe/1ZBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-16T18:31:11.683486Z","bundle_sha256":"9af7c6883d81700ca113f4370a60f39d55ef5c85b226e90350389bf1e82601eb"}}