{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:GILA6RH7KAAVPR6CT424VC6NBJ","short_pith_number":"pith:GILA6RH7","canonical_record":{"source":{"id":"2605.14036","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T18:56:02Z","cross_cats_sorted":["cs.CC","cs.CL","cs.LG"],"title_canon_sha256":"7bae97d102aa2a92bf707888b448dc0f6cab7cdb252106230e0e7c0c7b47607f","abstract_canon_sha256":"23cb108551527bdead40da023a647b74a5962a98c4906f5f4ef605cd482d0711"},"schema_version":"1.0"},"canonical_sha256":"32160f44ff500157c7c29f35ca8bcd0a53b5c3ae15cc79f3205aba972254c6db","source":{"kind":"arxiv","id":"2605.14036","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14036","created_at":"2026-05-17T23:39:12Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14036v1","created_at":"2026-05-17T23:39:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14036","created_at":"2026-05-17T23:39:12Z"},{"alias_kind":"pith_short_12","alias_value":"GILA6RH7KAAV","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"GILA6RH7KAAVPR6C","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"GILA6RH7","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:GILA6RH7KAAVPR6CT424VC6NBJ","target":"record","payload":{"canonical_record":{"source":{"id":"2605.14036","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T18:56:02Z","cross_cats_sorted":["cs.CC","cs.CL","cs.LG"],"title_canon_sha256":"7bae97d102aa2a92bf707888b448dc0f6cab7cdb252106230e0e7c0c7b47607f","abstract_canon_sha256":"23cb108551527bdead40da023a647b74a5962a98c4906f5f4ef605cd482d0711"},"schema_version":"1.0"},"canonical_sha256":"32160f44ff500157c7c29f35ca8bcd0a53b5c3ae15cc79f3205aba972254c6db","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:12.791072Z","signature_b64":"EI5hKNMmWsctQj/T1twB+sy5E67HA3yMiPdO34R650uCqYvtznoP780+WL7QAabglxQBAhjC++77wF537Tw5AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"32160f44ff500157c7c29f35ca8bcd0a53b5c3ae15cc79f3205aba972254c6db","last_reissued_at":"2026-05-17T23:39:12.790602Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:12.790602Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.14036","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":"2026-05-17T23:39:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PQiGjafxWuL7H7N8mfRN5QAI9tp0g3ALUNIfK/OPC9163qWl2z7+67j67+746GDfL50KGcQT5Xhx7ujbJN/YDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-19T11:51:22.247158Z"},"content_sha256":"a121b93a603668157c0b796676bdda98dc0e19c37123aa87ece6275c69230c75","schema_version":"1.0","event_id":"sha256:a121b93a603668157c0b796676bdda98dc0e19c37123aa87ece6275c69230c75"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:GILA6RH7KAAVPR6CT424VC6NBJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Enhanced and Efficient Reasoning in Large Learning Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Recoding data to Unary Relational Integracode lets large models learn relational rules in polynomial time.","cross_cats":["cs.CC","cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Leslie G. Valiant","submitted_at":"2026-05-13T18:56:02Z","abstract_excerpt":"In current Large Language Models we can trust the production of smoothly flowing prose on the basis of the principles of machine learning. However, there is no comparably principled basis to justify trust in the content of the text produced. It appears to be conventional wisdom that addressing this issue by adding more principled reasoning is not computationally affordable.\n  Here we propose a principled method of reasoning that is efficient enough to be practical for large language models. Further, the method allows the retention of much of the currently used software and hardware base. Our m"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"This recoding has the surprising and fortuitous property that, while succinct, it makes the task of learning a core subset of relational rules that hold in the world described in the training data polynomial time learnable in a defined sense, the polynomial depending on the complexity of the rule.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That preprocessing natural language text into Unary Relational Integracode can be performed efficiently and accurately enough to expose the relevant relationships without introducing prohibitive computational cost or information loss.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Preprocessing text into Unary Relational Integracode enables polynomial-time learning of relational rules for sound reasoning in large language models.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Recoding data to Unary Relational Integracode lets large models learn relational rules in polynomial time.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"f9ba3a1f56c92892976fe9c80256cf0ee9d818d66e820633a66c1d11bbcb4a0d"},"source":{"id":"2605.14036","kind":"arxiv","version":1},"verdict":{"id":"2dfdf569-6cec-4f4b-9796-88640ac4fba5","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T05:32:56.130682Z","strongest_claim":"This recoding has the surprising and fortuitous property that, while succinct, it makes the task of learning a core subset of relational rules that hold in the world described in the training data polynomial time learnable in a defined sense, the polynomial depending on the complexity of the rule.","one_line_summary":"Preprocessing text into Unary Relational Integracode enables polynomial-time learning of relational rules for sound reasoning in large language models.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That preprocessing natural language text into Unary Relational Integracode can be performed efficiently and accurately enough to expose the relevant relationships without introducing prohibitive computational cost or information loss.","pith_extraction_headline":"Recoding data to Unary Relational Integracode lets large models learn relational rules in polynomial time."},"references":{"count":40,"sample":[{"doi":"","year":1973,"title":"J. R. Anderson and G. H. Bower.Human Associative Memory. Psychology Press, New York, 1973","work_id":"e4c8dd95-824d-4504-9ab8-735ee4c81a62","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"B. Barak, B. Edelman, S. Goel, S. Kakade, E. Malach, and C. Zhang. Hidden progress in deep learning: Sgd learns parities near the computational limit. InAdvances in Neural Information Processing Syste","work_id":"83be1aa7-bd6d-4a44-9abf-515f0d0b5590","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"I. Beltagy, M. E. Peters, and A. Cohan. Longformer: The long-document transformer. arXiv preprint, 2020","work_id":"863d34aa-d706-4408-9976-0f38d2e3ebac","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"Choromanski and et al","work_id":"dee88f8d-7375-4703-9cbc-486bf44dd9bb","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"A. Daniely and G. Vardi. From local pseudorandom generators to hardness of learning. InConference on Learning Theory, pages 358–1394. PMLR, 2021","work_id":"31d2f96b-a7ea-432a-9361-ef98af63284b","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":40,"snapshot_sha256":"5561a5ab3c6311f4d94c65723848a8250a6678574b426f0f56f6b28a84f714a4","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":"2dfdf569-6cec-4f4b-9796-88640ac4fba5"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0rXW9POME5+rYq6hnmMviuo9op4rEUHiSrFYcek69Qt4JPMvVHsY+jiC0TMN9+3ThqY2YI9scj+PBLA7E5WGBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-19T11:51:22.247803Z"},"content_sha256":"7f196150403cba57b1d6e5cfa45c907b9a39863a61fbf02c2aaced4a718665cc","schema_version":"1.0","event_id":"sha256:7f196150403cba57b1d6e5cfa45c907b9a39863a61fbf02c2aaced4a718665cc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GILA6RH7KAAVPR6CT424VC6NBJ/bundle.json","state_url":"https://pith.science/pith/GILA6RH7KAAVPR6CT424VC6NBJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GILA6RH7KAAVPR6CT424VC6NBJ/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":"2026-05-19T11:51:22Z","links":{"resolver":"https://pith.science/pith/GILA6RH7KAAVPR6CT424VC6NBJ","bundle":"https://pith.science/pith/GILA6RH7KAAVPR6CT424VC6NBJ/bundle.json","state":"https://pith.science/pith/GILA6RH7KAAVPR6CT424VC6NBJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GILA6RH7KAAVPR6CT424VC6NBJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:GILA6RH7KAAVPR6CT424VC6NBJ","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":"23cb108551527bdead40da023a647b74a5962a98c4906f5f4ef605cd482d0711","cross_cats_sorted":["cs.CC","cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T18:56:02Z","title_canon_sha256":"7bae97d102aa2a92bf707888b448dc0f6cab7cdb252106230e0e7c0c7b47607f"},"schema_version":"1.0","source":{"id":"2605.14036","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14036","created_at":"2026-05-17T23:39:12Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14036v1","created_at":"2026-05-17T23:39:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14036","created_at":"2026-05-17T23:39:12Z"},{"alias_kind":"pith_short_12","alias_value":"GILA6RH7KAAV","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"GILA6RH7KAAVPR6C","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"GILA6RH7","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:7f196150403cba57b1d6e5cfa45c907b9a39863a61fbf02c2aaced4a718665cc","target":"graph","created_at":"2026-05-17T23:39:12Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"This recoding has the surprising and fortuitous property that, while succinct, it makes the task of learning a core subset of relational rules that hold in the world described in the training data polynomial time learnable in a defined sense, the polynomial depending on the complexity of the rule."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That preprocessing natural language text into Unary Relational Integracode can be performed efficiently and accurately enough to expose the relevant relationships without introducing prohibitive computational cost or information loss."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Preprocessing text into Unary Relational Integracode enables polynomial-time learning of relational rules for sound reasoning in large language models."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Recoding data to Unary Relational Integracode lets large models learn relational rules in polynomial time."}],"snapshot_sha256":"f9ba3a1f56c92892976fe9c80256cf0ee9d818d66e820633a66c1d11bbcb4a0d"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"In current Large Language Models we can trust the production of smoothly flowing prose on the basis of the principles of machine learning. However, there is no comparably principled basis to justify trust in the content of the text produced. It appears to be conventional wisdom that addressing this issue by adding more principled reasoning is not computationally affordable.\n  Here we propose a principled method of reasoning that is efficient enough to be practical for large language models. Further, the method allows the retention of much of the currently used software and hardware base. Our m","authors_text":"Leslie G. Valiant","cross_cats":["cs.CC","cs.CL","cs.LG"],"headline":"Recoding data to Unary Relational Integracode lets large models learn relational rules in polynomial time.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T18:56:02Z","title":"Enhanced and Efficient Reasoning in Large Learning Models"},"references":{"count":40,"internal_anchors":0,"resolved_work":40,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"J. R. Anderson and G. H. Bower.Human Associative Memory. Psychology Press, New York, 1973","work_id":"e4c8dd95-824d-4504-9ab8-735ee4c81a62","year":1973},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"B. Barak, B. Edelman, S. Goel, S. Kakade, E. Malach, and C. Zhang. Hidden progress in deep learning: Sgd learns parities near the computational limit. InAdvances in Neural Information Processing Syste","work_id":"83be1aa7-bd6d-4a44-9abf-515f0d0b5590","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"I. Beltagy, M. E. Peters, and A. Cohan. Longformer: The long-document transformer. arXiv preprint, 2020","work_id":"863d34aa-d706-4408-9976-0f38d2e3ebac","year":2020},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Choromanski and et al","work_id":"dee88f8d-7375-4703-9cbc-486bf44dd9bb","year":2020},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"A. Daniely and G. Vardi. From local pseudorandom generators to hardness of learning. InConference on Learning Theory, pages 358–1394. PMLR, 2021","work_id":"31d2f96b-a7ea-432a-9361-ef98af63284b","year":2021}],"snapshot_sha256":"5561a5ab3c6311f4d94c65723848a8250a6678574b426f0f56f6b28a84f714a4"},"source":{"id":"2605.14036","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T05:32:56.130682Z","id":"2dfdf569-6cec-4f4b-9796-88640ac4fba5","model_set":{"reader":"grok-4.3"},"one_line_summary":"Preprocessing text into Unary Relational Integracode enables polynomial-time learning of relational rules for sound reasoning in large language models.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Recoding data to Unary Relational Integracode lets large models learn relational rules in polynomial time.","strongest_claim":"This recoding has the surprising and fortuitous property that, while succinct, it makes the task of learning a core subset of relational rules that hold in the world described in the training data polynomial time learnable in a defined sense, the polynomial depending on the complexity of the rule.","weakest_assumption":"That preprocessing natural language text into Unary Relational Integracode can be performed efficiently and accurately enough to expose the relevant relationships without introducing prohibitive computational cost or information loss."}},"verdict_id":"2dfdf569-6cec-4f4b-9796-88640ac4fba5"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:a121b93a603668157c0b796676bdda98dc0e19c37123aa87ece6275c69230c75","target":"record","created_at":"2026-05-17T23:39:12Z","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":"23cb108551527bdead40da023a647b74a5962a98c4906f5f4ef605cd482d0711","cross_cats_sorted":["cs.CC","cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T18:56:02Z","title_canon_sha256":"7bae97d102aa2a92bf707888b448dc0f6cab7cdb252106230e0e7c0c7b47607f"},"schema_version":"1.0","source":{"id":"2605.14036","kind":"arxiv","version":1}},"canonical_sha256":"32160f44ff500157c7c29f35ca8bcd0a53b5c3ae15cc79f3205aba972254c6db","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"32160f44ff500157c7c29f35ca8bcd0a53b5c3ae15cc79f3205aba972254c6db","first_computed_at":"2026-05-17T23:39:12.790602Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:12.790602Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"EI5hKNMmWsctQj/T1twB+sy5E67HA3yMiPdO34R650uCqYvtznoP780+WL7QAabglxQBAhjC++77wF537Tw5AQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:12.791072Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.14036","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a121b93a603668157c0b796676bdda98dc0e19c37123aa87ece6275c69230c75","sha256:7f196150403cba57b1d6e5cfa45c907b9a39863a61fbf02c2aaced4a718665cc"],"state_sha256":"fd6980747e34c66be2a548b2f387e741d5ad2e405c89d453f306fa57ad2972b8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3wtVeMrMtLjEjD21UfQousp5QVXao1DCxtwjzpzjvEP48pvgyC4AGi+VIrPlGvfd4AH2v0K7Cz1EoiB85yvwCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-19T11:51:22.249945Z","bundle_sha256":"f712e4e5432e034e0ebcb057dd5bb5e9dfaf0ea95d130608ddd8ed7fd87b1bcc"}}