{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VWPP6L7NKA7FHVUHQJQMHUEEVJ","short_pith_number":"pith:VWPP6L7N","schema_version":"1.0","canonical_sha256":"ad9eff2fed503e53d6878260c3d084aa6e9f5f3ca455e75f383e9922941e2e30","source":{"kind":"arxiv","id":"2510.26792","version":2},"attestation_state":"computed","paper":{"title":"Learning Pseudorandom Numbers with Transformers: Permuted Congruential Generators, Curricula, and Interpretability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.dis-nn","cs.CR"],"primary_cat":"cs.LG","authors_text":"Maissam Barkeshli, Tao Tao","submitted_at":"2025-10-30T17:59:09Z","abstract_excerpt":"We study the ability of Transformer models to learn sequences generated by Permuted Congruential Generators (PCGs), a widely used family of pseudo-random number generators (PRNGs). PCGs introduce substantial additional difficulty over linear congruential generators (LCGs) by applying a series of bit-wise shifts, XORs, rotations and truncations to the hidden state. We show that Transformers can nevertheless successfully perform in-context prediction on unseen sequences from diverse PCG variants, in tasks that are beyond published classical attacks. In our experiments we scale moduli up to $2^{2"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2510.26792","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-10-30T17:59:09Z","cross_cats_sorted":["cond-mat.dis-nn","cs.CR"],"title_canon_sha256":"3e9b9e020c969c29fb73ff23adaba9cdaa6f091126a5e08bf397a20e8b8c3ac5","abstract_canon_sha256":"8e71f36a6cc0e3b3a9eba79ef66966cf082f40ae2418ebf225f0ea27068b59ca"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-17T00:20:41.609900Z","signature_b64":"IkuiI9v/W5K+8fOiXVDiyb+3d8Ecdt9ZDohMMTKsTmNUFsKx31RLVgrgSlnc+nxjE5g3xrEEjBxM+sc0FVpjDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ad9eff2fed503e53d6878260c3d084aa6e9f5f3ca455e75f383e9922941e2e30","last_reissued_at":"2026-07-17T00:20:41.608980Z","signature_status":"signed_v1","first_computed_at":"2026-07-17T00:20:41.608980Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning Pseudorandom Numbers with Transformers: Permuted Congruential Generators, Curricula, and Interpretability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.dis-nn","cs.CR"],"primary_cat":"cs.LG","authors_text":"Maissam Barkeshli, Tao Tao","submitted_at":"2025-10-30T17:59:09Z","abstract_excerpt":"We study the ability of Transformer models to learn sequences generated by Permuted Congruential Generators (PCGs), a widely used family of pseudo-random number generators (PRNGs). PCGs introduce substantial additional difficulty over linear congruential generators (LCGs) by applying a series of bit-wise shifts, XORs, rotations and truncations to the hidden state. We show that Transformers can nevertheless successfully perform in-context prediction on unseen sequences from diverse PCG variants, in tasks that are beyond published classical attacks. In our experiments we scale moduli up to $2^{2"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.26792","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/2510.26792/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2510.26792","created_at":"2026-07-17T00:20:41.609413+00:00"},{"alias_kind":"arxiv_version","alias_value":"2510.26792v2","created_at":"2026-07-17T00:20:41.609413+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.26792","created_at":"2026-07-17T00:20:41.609413+00:00"},{"alias_kind":"pith_short_12","alias_value":"VWPP6L7NKA7F","created_at":"2026-07-17T00:20:41.609413+00:00"},{"alias_kind":"pith_short_16","alias_value":"VWPP6L7NKA7FHVUH","created_at":"2026-07-17T00:20:41.609413+00:00"},{"alias_kind":"pith_short_8","alias_value":"VWPP6L7N","created_at":"2026-07-17T00:20:41.609413+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VWPP6L7NKA7FHVUHQJQMHUEEVJ","json":"https://pith.science/pith/VWPP6L7NKA7FHVUHQJQMHUEEVJ.json","graph_json":"https://pith.science/api/pith-number/VWPP6L7NKA7FHVUHQJQMHUEEVJ/graph.json","events_json":"https://pith.science/api/pith-number/VWPP6L7NKA7FHVUHQJQMHUEEVJ/events.json","paper":"https://pith.science/paper/VWPP6L7N"},"agent_actions":{"view_html":"https://pith.science/pith/VWPP6L7NKA7FHVUHQJQMHUEEVJ","download_json":"https://pith.science/pith/VWPP6L7NKA7FHVUHQJQMHUEEVJ.json","view_paper":"https://pith.science/paper/VWPP6L7N","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2510.26792&json=true","fetch_graph":"https://pith.science/api/pith-number/VWPP6L7NKA7FHVUHQJQMHUEEVJ/graph.json","fetch_events":"https://pith.science/api/pith-number/VWPP6L7NKA7FHVUHQJQMHUEEVJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VWPP6L7NKA7FHVUHQJQMHUEEVJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VWPP6L7NKA7FHVUHQJQMHUEEVJ/action/storage_attestation","attest_author":"https://pith.science/pith/VWPP6L7NKA7FHVUHQJQMHUEEVJ/action/author_attestation","sign_citation":"https://pith.science/pith/VWPP6L7NKA7FHVUHQJQMHUEEVJ/action/citation_signature","submit_replication":"https://pith.science/pith/VWPP6L7NKA7FHVUHQJQMHUEEVJ/action/replication_record"}},"created_at":"2026-07-17T00:20:41.609413+00:00","updated_at":"2026-07-17T00:20:41.609413+00:00"}