{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:YOM5ZNLWAM6OBBC5PZEXJHFDAA","short_pith_number":"pith:YOM5ZNLW","schema_version":"1.0","canonical_sha256":"c399dcb576033ce0845d7e49749ca300058cace38bb4eed58ff43cbae1aa1ad4","source":{"kind":"arxiv","id":"2309.08632","version":1},"attestation_state":"computed","paper":{"title":"Pretraining on the Test Set Is All You Need","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Rylan Schaeffer","submitted_at":"2023-09-13T19:47:33Z","abstract_excerpt":"Inspired by recent work demonstrating the promise of smaller Transformer-based language models pretrained on carefully curated data, we supercharge such approaches by investing heavily in curating a novel, high quality, non-synthetic data mixture based solely on evaluation benchmarks. Using our novel dataset mixture consisting of less than 100 thousand tokens, we pretrain a 1 million parameter transformer-based LLM \\textbf{phi-CTNL} (pronounced ``fictional\") that achieves perfect results across diverse academic benchmarks, strictly outperforming all known foundation models. \\textbf{phi-CTNL} a"},"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":"2309.08632","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2023-09-13T19:47:33Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"0b2b62909a29c37c435a5c62f4fa21c86859920143bae7e1a9085e3b3925bcf6","abstract_canon_sha256":"1a2fb14901e992b8b4eda218ab636cc18f7b5f3cb2bbc8eff19617951a965dfe"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:51:11.073778Z","signature_b64":"u0JTwwBn/wOM2l3C/ENnLHZQNZ7Qb3cUhq27NJu40hChEWKui1JlpziWxM5hTK/K0pdK8kVvZqBKHMZNQ7E+CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c399dcb576033ce0845d7e49749ca300058cace38bb4eed58ff43cbae1aa1ad4","last_reissued_at":"2026-07-05T06:51:11.073335Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:51:11.073335Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Pretraining on the Test Set Is All You Need","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Rylan Schaeffer","submitted_at":"2023-09-13T19:47:33Z","abstract_excerpt":"Inspired by recent work demonstrating the promise of smaller Transformer-based language models pretrained on carefully curated data, we supercharge such approaches by investing heavily in curating a novel, high quality, non-synthetic data mixture based solely on evaluation benchmarks. Using our novel dataset mixture consisting of less than 100 thousand tokens, we pretrain a 1 million parameter transformer-based LLM \\textbf{phi-CTNL} (pronounced ``fictional\") that achieves perfect results across diverse academic benchmarks, strictly outperforming all known foundation models. \\textbf{phi-CTNL} a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.08632","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/2309.08632/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":"2309.08632","created_at":"2026-07-05T06:51:11.073393+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.08632v1","created_at":"2026-07-05T06:51:11.073393+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.08632","created_at":"2026-07-05T06:51:11.073393+00:00"},{"alias_kind":"pith_short_12","alias_value":"YOM5ZNLWAM6O","created_at":"2026-07-05T06:51:11.073393+00:00"},{"alias_kind":"pith_short_16","alias_value":"YOM5ZNLWAM6OBBC5","created_at":"2026-07-05T06:51:11.073393+00:00"},{"alias_kind":"pith_short_8","alias_value":"YOM5ZNLW","created_at":"2026-07-05T06:51:11.073393+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.08904","citing_title":"OPT-BENCH: Evaluating the Iterative Self-Optimization of LLM Agents in Large-Scale Search Spaces","ref_index":74,"is_internal_anchor":false},{"citing_arxiv_id":"2405.09818","citing_title":"Chameleon: Mixed-Modal Early-Fusion Foundation Models","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YOM5ZNLWAM6OBBC5PZEXJHFDAA","json":"https://pith.science/pith/YOM5ZNLWAM6OBBC5PZEXJHFDAA.json","graph_json":"https://pith.science/api/pith-number/YOM5ZNLWAM6OBBC5PZEXJHFDAA/graph.json","events_json":"https://pith.science/api/pith-number/YOM5ZNLWAM6OBBC5PZEXJHFDAA/events.json","paper":"https://pith.science/paper/YOM5ZNLW"},"agent_actions":{"view_html":"https://pith.science/pith/YOM5ZNLWAM6OBBC5PZEXJHFDAA","download_json":"https://pith.science/pith/YOM5ZNLWAM6OBBC5PZEXJHFDAA.json","view_paper":"https://pith.science/paper/YOM5ZNLW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.08632&json=true","fetch_graph":"https://pith.science/api/pith-number/YOM5ZNLWAM6OBBC5PZEXJHFDAA/graph.json","fetch_events":"https://pith.science/api/pith-number/YOM5ZNLWAM6OBBC5PZEXJHFDAA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YOM5ZNLWAM6OBBC5PZEXJHFDAA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YOM5ZNLWAM6OBBC5PZEXJHFDAA/action/storage_attestation","attest_author":"https://pith.science/pith/YOM5ZNLWAM6OBBC5PZEXJHFDAA/action/author_attestation","sign_citation":"https://pith.science/pith/YOM5ZNLWAM6OBBC5PZEXJHFDAA/action/citation_signature","submit_replication":"https://pith.science/pith/YOM5ZNLWAM6OBBC5PZEXJHFDAA/action/replication_record"}},"created_at":"2026-07-05T06:51:11.073393+00:00","updated_at":"2026-07-05T06:51:11.073393+00:00"}