{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:RKJAHZERU6TEB6PYJ63BIRCCZG","short_pith_number":"pith:RKJAHZER","schema_version":"1.0","canonical_sha256":"8a9203e491a7a640f9f84fb6144442c9a65cf87ec8aa3aca99876991ccb6c42d","source":{"kind":"arxiv","id":"2312.09241","version":1},"attestation_state":"computed","paper":{"title":"TinyGSM: achieving >80% on GSM8k with small language models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Anh Nguyen, Bingbin Liu, Janardhan Kulkarni, Rachel Ward, Ronen Eldan, Sebastien Bubeck, Yi Zhang, Yuanzhi Li","submitted_at":"2023-12-14T18:58:28Z","abstract_excerpt":"Small-scale models offer various computational advantages, and yet to which extent size is critical for problem-solving abilities remains an open question. Specifically for solving grade school math, the smallest model size so far required to break the 80\\% barrier on the GSM8K benchmark remains to be 34B. Our work studies how high-quality datasets may be the key for small language models to acquire mathematical reasoning. We introduce \\texttt{TinyGSM}, a synthetic dataset of 12.3M grade school math problems paired with Python solutions, generated fully by GPT-3.5. After finetuning on \\texttt{"},"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":"2312.09241","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-12-14T18:58:28Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"dc4e98083d1ef0117b7fdc67d69b770ff110738adaeb1a2621b14090041041c0","abstract_canon_sha256":"ce5ddcf8a2638e32314e672ca938209f0c8459c207f55e4ba4c47b3922fba6b4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:24:16.376672Z","signature_b64":"XbXQv4ozB8U5QXVqAxwK3dFgWciQo/xDUxo+d0MPS/akOMmgwX8nZR1RIWCHgjY3TAbTqGWjQGPld7WnfME5DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8a9203e491a7a640f9f84fb6144442c9a65cf87ec8aa3aca99876991ccb6c42d","last_reissued_at":"2026-07-05T07:24:16.376186Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:24:16.376186Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TinyGSM: achieving >80% on GSM8k with small language models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Anh Nguyen, Bingbin Liu, Janardhan Kulkarni, Rachel Ward, Ronen Eldan, Sebastien Bubeck, Yi Zhang, Yuanzhi Li","submitted_at":"2023-12-14T18:58:28Z","abstract_excerpt":"Small-scale models offer various computational advantages, and yet to which extent size is critical for problem-solving abilities remains an open question. Specifically for solving grade school math, the smallest model size so far required to break the 80\\% barrier on the GSM8K benchmark remains to be 34B. Our work studies how high-quality datasets may be the key for small language models to acquire mathematical reasoning. We introduce \\texttt{TinyGSM}, a synthetic dataset of 12.3M grade school math problems paired with Python solutions, generated fully by GPT-3.5. After finetuning on \\texttt{"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.09241","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/2312.09241/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":"2312.09241","created_at":"2026-07-05T07:24:16.376245+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.09241v1","created_at":"2026-07-05T07:24:16.376245+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.09241","created_at":"2026-07-05T07:24:16.376245+00:00"},{"alias_kind":"pith_short_12","alias_value":"RKJAHZERU6TE","created_at":"2026-07-05T07:24:16.376245+00:00"},{"alias_kind":"pith_short_16","alias_value":"RKJAHZERU6TEB6PY","created_at":"2026-07-05T07:24:16.376245+00:00"},{"alias_kind":"pith_short_8","alias_value":"RKJAHZER","created_at":"2026-07-05T07:24:16.376245+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24773","citing_title":"Posterior Refinement: Fast Language Generation via Any-Order Flow Maps","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2606.21627","citing_title":"Counsel: A Meta-Evaluation Dataset for Agentic Tasks","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12385","citing_title":"Which Models Are Our Models Built On? Auditing Invisible Dependencies in Modern LLMs","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2606.02241","citing_title":"BlockGen: Flexible Blockwise Sequence Modeling with Hybrid Samplers","ref_index":138,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26106","citing_title":"Looped Diffusion Language Models","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26496","citing_title":"Dense2MoE: Pushing the Pareto Frontier of On-Device LLMs via Unified Pruning and Upcycling","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00683","citing_title":"OCC-RAG: Optimal Cognitive Core for Faithful Question Answering","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2501.05465","citing_title":"Small Language Models (SLMs) Can Still Pack a Punch: A survey (updated 2026)","ref_index":74,"is_internal_anchor":false},{"citing_arxiv_id":"2602.08167","citing_title":"Self-Supervised Bootstrapping of Action-Predictive Embodied Reasoning","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2510.04265","citing_title":"Don't Pass@k: A Bayesian Framework for Large Language Model Evaluation","ref_index":52,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RKJAHZERU6TEB6PYJ63BIRCCZG","json":"https://pith.science/pith/RKJAHZERU6TEB6PYJ63BIRCCZG.json","graph_json":"https://pith.science/api/pith-number/RKJAHZERU6TEB6PYJ63BIRCCZG/graph.json","events_json":"https://pith.science/api/pith-number/RKJAHZERU6TEB6PYJ63BIRCCZG/events.json","paper":"https://pith.science/paper/RKJAHZER"},"agent_actions":{"view_html":"https://pith.science/pith/RKJAHZERU6TEB6PYJ63BIRCCZG","download_json":"https://pith.science/pith/RKJAHZERU6TEB6PYJ63BIRCCZG.json","view_paper":"https://pith.science/paper/RKJAHZER","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.09241&json=true","fetch_graph":"https://pith.science/api/pith-number/RKJAHZERU6TEB6PYJ63BIRCCZG/graph.json","fetch_events":"https://pith.science/api/pith-number/RKJAHZERU6TEB6PYJ63BIRCCZG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RKJAHZERU6TEB6PYJ63BIRCCZG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RKJAHZERU6TEB6PYJ63BIRCCZG/action/storage_attestation","attest_author":"https://pith.science/pith/RKJAHZERU6TEB6PYJ63BIRCCZG/action/author_attestation","sign_citation":"https://pith.science/pith/RKJAHZERU6TEB6PYJ63BIRCCZG/action/citation_signature","submit_replication":"https://pith.science/pith/RKJAHZERU6TEB6PYJ63BIRCCZG/action/replication_record"}},"created_at":"2026-07-05T07:24:16.376245+00:00","updated_at":"2026-07-05T07:24:16.376245+00:00"}