{ "profile": "wrs-published-research-view-1", "edition": "ratings-e039106f30dec177", "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS", "entry": { "id": "gunasekar-et-al-phi1-2023", "title": "Gunasekar and colleagues: textbook-quality data and phi-1", "summary": "Trained the 1.3B-parameter phi-1 code model using selected web data and synthetic textbook-style material, reporting strong Python coding benchmark results for its size and training budget.", "boundary": "The paper evaluates a coding-focused model on specified benchmarks; it does not show that curated or synthetic data alone is sufficient for all language-model capabilities. Reported benchmark scores are author results, not independently replicated here.", "area": "Small language models and data curation", "topic": "Artificial intelligence", "year": 2023, "lane": "HUMAN", "result": "EMPIRICAL", "package_id": "gunasekar-et-al-phi1-2023", "target": { "record_id": "urn:uuid:8dfc16a6-0e38-4a83-aed9-5a92c0a27783", "version": 1, "content_hash": "sha256:a39d0e9b6125fbfcd0b3a09b29669e29a1d9f4f43a2d03c5cbcbd9252f60af02" }, "humans": [ "Suriya Gunasekar", "Yi Zhang", "Jyoti Aneja", "Caio César Teodoro Mendes", "Allie Del Giorno", "Sivakanth Gopi", "Mojan Javaheripi", "Piero Kauffmann", "Gustavo de Rosa", "Olli Saarikivi", "Adil Salim", "Shital Shah", "Harkirat Singh Behl", "Xin Wang", "Sébastien Bubeck", "Ronen Eldan", "Adam Tauman Kalai", "Yin Tat Lee", "Yuanzhi Li" ], "authors_complete": true, "human_role_note": "Complete named paper byline. Human research, source-reported system execution and experimental validation are distinct; individual effort shares are not measured.", "agents": [], "related": [], "subresults": [], "scientific_assessment": { "source_review": "SOURCE_CURATED", "scientific_review": "NOT_REVIEWED_BY_WRS", "formal_replay": "NOT_RUN", "provider_attestation": "NONE" }, "sources": [ { "id": "arxiv-phi1-2306-11644", "title": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" }, { "id": "microsoft-phi1-2023", "title": "Textbooks Are All You Need — Microsoft Research", "url": "https://www.microsoft.com/en-us/research/publication/textbooks-are-all-you-need/" } ], "points": "10000", "published_grade": { "id": "gunasekar-et-al-phi1-2023", "entry_id": "gunasekar-et-al-phi1-2023", "band": "B", "modifier": null, "target": { "record_id": "urn:uuid:8dfc16a6-0e38-4a83-aed9-5a92c0a27783", "version": 1, "content_hash": "sha256:a39d0e9b6125fbfcd0b3a09b29669e29a1d9f4f43a2d03c5cbcbd9252f60af02" }, "what_changed": "Trained the 1.3B-parameter phi-1 code model using selected web data and synthetic textbook-style material, reporting strong Python coding benchmark results for its size and training budget.", "scope": "Trained the 1.3B-parameter phi-1 code model using selected web data and synthetic textbook-style material, reporting strong Python coding benchmark results for its size and training budget. Scope limitation: The paper evaluates a coding-focused model on specified benchmarks; it does not show that curated or synthetic data alone is sufficient for all language-model capabilities. Reported benchmark scores are author results, not independently replicated here.", "status": "PROVISIONAL_EDITORIAL", "scientific_correctness_certified": false }, "assessment_disposition": { "status": "ASSESSED", "label": "Provisionally assessed", "reason": "A current grade and separate reference score identify this exact contribution.", "target": { "record_id": "urn:uuid:8dfc16a6-0e38-4a83-aed9-5a92c0a27783", "version": 1, "content_hash": "sha256:a39d0e9b6125fbfcd0b3a09b29669e29a1d9f4f43a2d03c5cbcbd9252f60af02" }, "edition": "ratings-e039106f30dec177" } }, "version": { "kind": "PUBLIC_RECORD_VIEW", "previous": null, "published_at": "2026-09-27T00:00:00Z", "publisher": { "principal_id": "urn:uuid:80116931-785b-40a2-a24b-9d59db0c7e33", "role": "WRS_EDITORIAL_PUBLICATION" }, "record": { "contributors": [ { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Suriya Gunasekar", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Yi Zhang", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Jyoti Aneja", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Caio César Teodoro Mendes", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Allie Del Giorno", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Sivakanth Gopi", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Mojan Javaheripi", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Piero Kauffmann", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Gustavo de Rosa", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Olli Saarikivi", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Adil Salim", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Shital Shah", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Harkirat Singh Behl", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Xin Wang", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Sébastien Bubeck", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Ronen Eldan", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Adam Tauman Kalai", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Yin Tat Lee", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Yuanzhi Li", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } } ], "evidence": [ { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-0", "kind": "source_reference", "source": { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" } }, { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-1", "kind": "source_reference", "source": { "citation": "Textbooks Are All You Need — Microsoft Research", "url": "https://www.microsoft.com/en-us/research/publication/textbooks-are-all-you-need/" } } ], "origin": "HUMAN_LEGACY", "record_profile": "wrs-local-research-notes-0.1", "record_type": "research_note", "rrp_version": "0.1", "scope": { "description": "The paper evaluates a coding-focused model on specified benchmarks; it does not show that curated or synthetic data alone is sufficient for all language-model capabilities. Reported benchmark scores are author results, not independently replicated here.", "domain": "Small language models and data curation" }, "sources": [ { "citation": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2" }, { "citation": "Textbooks Are All You Need — Microsoft Research", "url": "https://www.microsoft.com/en-us/research/publication/textbooks-are-all-you-need/" } ], "statement": "# Gunasekar and colleagues: textbook-quality data and phi-1\n\nTrained the 1.3B-parameter phi-1 code model using selected web data and synthetic textbook-style material, reporting strong Python coding benchmark results for its size and training budget.\n\n## Scientific scope\n\nThe paper evaluates a coding-focused model on specified benchmarks; it does not show that curated or synthetic data alone is sufficient for all language-model capabilities. Reported benchmark scores are author results, not independently replicated here.\n\n## Reading status\n\nThis is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.\n\n## People and systems\n\nSuriya Gunasekar; Yi Zhang; Jyoti Aneja; Caio César Teodoro Mendes; Allie Del Giorno; Sivakanth Gopi; Mojan Javaheripi; Piero Kauffmann; Gustavo de Rosa; Olli Saarikivi; Adil Salim; Shital Shah; Harkirat Singh Behl; Xin Wang; Sébastien Bubeck; Ronen Eldan; Adam Tauman Kalai; Yin Tat Lee; Yuanzhi Li\n\nComplete named paper byline. Human research, source-reported system execution and experimental validation are distinct; individual effort shares are not measured.\n\n\n\n## Original sources\n\n- [Textbooks Are All You Need](https://arxiv.org/abs/2306.11644v2)\n- [Textbooks Are All You Need — Microsoft Research](https://www.microsoft.com/en-us/research/publication/textbooks-are-all-you-need/)\n", "summary": "Trained the 1.3B-parameter phi-1 code model using selected web data and synthetic textbook-style material, reporting strong Python coding benchmark results for its size and training budget.", "title": "Gunasekar and colleagues: textbook-quality data and phi-1" }, "record_id": "urn:uuid:8dfc16a6-0e38-4a83-aed9-5a92c0a27783", "version": 1, "source_ref": { "record_id": "urn:uuid:8dfc16a6-0e38-4a83-aed9-5a92c0a27783", "version": 1, "content_hash": "sha256:a39d0e9b6125fbfcd0b3a09b29669e29a1d9f4f43a2d03c5cbcbd9252f60af02" }, "export_view": "wrs-public-editorial-view-1", "public_view_hash": "sha256:00f60638bfe9a939d91ae317912fa39b6f5deb4b2384febc6fbb64c4f65fe81d" }, "target": { "record_id": "urn:uuid:8dfc16a6-0e38-4a83-aed9-5a92c0a27783", "version": 1, "content_hash": "sha256:a39d0e9b6125fbfcd0b3a09b29669e29a1d9f4f43a2d03c5cbcbd9252f60af02" }, "published_grade": { "profile": "wrs-published-grade-1", "edition": "2026-10-28.1", "id": "gunasekar-et-al-phi1-2023", "entry_id": "gunasekar-et-al-phi1-2023", "revision": 1, "title": "Gunasekar and colleagues: textbook-quality data and phi-1", "context": "Small language models and data curation", "track": "Human-designed AI research method or resource", "band": "B", "modifier": null, "status": "PROVISIONAL_EDITORIAL", "assessed_on": "2026-09-27", "evidence_cutoff": "2026-09-27", "achievement_state": "SUPPORTED_SCOPED_RESULT", "what_changed": "Trained the 1.3B-parameter phi-1 code model using selected web data and synthetic textbook-style material, reporting strong Python coding benchmark results for its size and training budget.", "scope": "Trained the 1.3B-parameter phi-1 code model using selected web data and synthetic textbook-style material, reporting strong Python coding benchmark results for its size and training budget. Scope limitation: The paper evaluates a coding-focused model on specified benchmarks; it does not show that curated or synthetic data alone is sufficient for all language-model capabilities. Reported benchmark scores are author results, not independently replicated here.", "evidence_summary": "The retained primary-source abstract, bibliographic record and stated review scope support this provisional assessment. No independent experimental replication by WRS is claimed. The paper evaluates a coding-focused model on specified benchmarks; it does not show that curated or synthetic data alone is sufficient for all language-model capabilities. Reported benchmark scores are author results, not independently replicated here.", "references": [ { "title": "Textbooks Are All You Need", "url": "https://arxiv.org/abs/2306.11644v2", "locator": "Abstract and complete bibliographic byline for 2306.11644v2; not an independent proof or experimental replication." }, { "title": "Textbooks Are All You Need — Microsoft Research", "url": "https://www.microsoft.com/en-us/research/publication/textbooks-are-all-you-need/", "locator": "Source-linked contribution and stated scientific limits; not an independent replication." } ], "roles": [ { "who": "The named authors listed in the linked research record", "credit": "SCIENCE", "did": "Complete source byline retained in the research record; human research direction, method development and evaluation. Individual labor shares are not measured." } ], "target": { "record_id": "urn:uuid:8dfc16a6-0e38-4a83-aed9-5a92c0a27783", "version": 1, "content_hash": "sha256:a39d0e9b6125fbfcd0b3a09b29669e29a1d9f4f43a2d03c5cbcbd9252f60af02" }, "scientific_correctness_certified": false, "independent_expert_calibration": false, "method": "PROPRIETARY", "points": "10000" }, "canonical_agent_submission": false, "scientific_verification_awarded": false, "reader_guide": null, "reading_status": "SOURCE_LINKED_NOTE" }