{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:GXPPOUBJLFAMUZXA4RTSK74EAG","short_pith_number":"pith:GXPPOUBJ","schema_version":"1.0","canonical_sha256":"35def750295940ca66e0e467257f8401ba0a69de5620a7921cc27a96ddeb6491","source":{"kind":"arxiv","id":"2212.10071","version":2},"attestation_state":"computed","paper":{"title":"Large Language Models Are Reasoning Teachers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Laura Schmid, Namgyu Ho, Se-Young Yun","submitted_at":"2022-12-20T08:24:45Z","abstract_excerpt":"Recent works have shown that chain-of-thought (CoT) prompting can elicit language models to solve complex reasoning tasks, step-by-step. However, prompt-based CoT methods are dependent on very large models such as GPT-3 175B which are prohibitive to deploy at scale. In this paper, we use these large models as reasoning teachers to enable complex reasoning in smaller models and reduce model size requirements by several orders of magnitude. We propose Fine-tune-CoT, a method that generates reasoning samples from very large teacher models to fine-tune smaller models. We evaluate our method on a w"},"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":"2212.10071","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-20T08:24:45Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"1062a580611f2c2d19cce8d3220d22daaa55d158fc127eed9e154dde8e8b8109","abstract_canon_sha256":"bd23db7a229244a0a2bf6bb35d54e5e70a307abe0bf26e40ce1bc807540db8eb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:19:54.711846Z","signature_b64":"2RJ5DDDN6Y7BNSQexw0iL02S5zWO6X015m8Fh7pvhLaTLTkE5e8RHoGNtl+NbPyKCj3JjJSJcVXPrAgf0sVDAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"35def750295940ca66e0e467257f8401ba0a69de5620a7921cc27a96ddeb6491","last_reissued_at":"2026-07-05T06:19:54.711425Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:19:54.711425Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Large Language Models Are Reasoning Teachers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Laura Schmid, Namgyu Ho, Se-Young Yun","submitted_at":"2022-12-20T08:24:45Z","abstract_excerpt":"Recent works have shown that chain-of-thought (CoT) prompting can elicit language models to solve complex reasoning tasks, step-by-step. However, prompt-based CoT methods are dependent on very large models such as GPT-3 175B which are prohibitive to deploy at scale. In this paper, we use these large models as reasoning teachers to enable complex reasoning in smaller models and reduce model size requirements by several orders of magnitude. We propose Fine-tune-CoT, a method that generates reasoning samples from very large teacher models to fine-tune smaller models. We evaluate our method on a w"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.10071","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/2212.10071/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":"2212.10071","created_at":"2026-07-05T06:19:54.711481+00:00"},{"alias_kind":"arxiv_version","alias_value":"2212.10071v2","created_at":"2026-07-05T06:19:54.711481+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.10071","created_at":"2026-07-05T06:19:54.711481+00:00"},{"alias_kind":"pith_short_12","alias_value":"GXPPOUBJLFAM","created_at":"2026-07-05T06:19:54.711481+00:00"},{"alias_kind":"pith_short_16","alias_value":"GXPPOUBJLFAMUZXA","created_at":"2026-07-05T06:19:54.711481+00:00"},{"alias_kind":"pith_short_8","alias_value":"GXPPOUBJ","created_at":"2026-07-05T06:19:54.711481+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.31048","citing_title":"Knowledge Distillation from Large Reasoning Models to Compact Student Models: A Case Study on the John O Bryan Mathematics Competition","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09881","citing_title":"Toward Calibrated, Fair, and accurate Deepfake Detection","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2504.02181","citing_title":"A Survey of Scaling in Large Language Model Reasoning","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2305.02301","citing_title":"Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes","ref_index":77,"is_internal_anchor":false},{"citing_arxiv_id":"2304.05376","citing_title":"ChemCrow: Augmenting large-language models with chemistry tools","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2404.14294","citing_title":"A Survey on Efficient Inference for Large Language Models","ref_index":124,"is_internal_anchor":false},{"citing_arxiv_id":"2303.17760","citing_title":"CAMEL: Communicative Agents for \"Mind\" Exploration of Large Language Model Society","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27233","citing_title":"Reinforced Agent: Inference-Time Feedback for Tool-Calling Agents","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05893","citing_title":"Logic-Regularized Verifier Elicits Reasoning from LLMs","ref_index":74,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07307","citing_title":"Rethinking Dense Sequential Chains: Reasoning Language Models Can Extract Answers from Sparse, Order-Shuffling Chain-of-Thoughts","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GXPPOUBJLFAMUZXA4RTSK74EAG","json":"https://pith.science/pith/GXPPOUBJLFAMUZXA4RTSK74EAG.json","graph_json":"https://pith.science/api/pith-number/GXPPOUBJLFAMUZXA4RTSK74EAG/graph.json","events_json":"https://pith.science/api/pith-number/GXPPOUBJLFAMUZXA4RTSK74EAG/events.json","paper":"https://pith.science/paper/GXPPOUBJ"},"agent_actions":{"view_html":"https://pith.science/pith/GXPPOUBJLFAMUZXA4RTSK74EAG","download_json":"https://pith.science/pith/GXPPOUBJLFAMUZXA4RTSK74EAG.json","view_paper":"https://pith.science/paper/GXPPOUBJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2212.10071&json=true","fetch_graph":"https://pith.science/api/pith-number/GXPPOUBJLFAMUZXA4RTSK74EAG/graph.json","fetch_events":"https://pith.science/api/pith-number/GXPPOUBJLFAMUZXA4RTSK74EAG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GXPPOUBJLFAMUZXA4RTSK74EAG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GXPPOUBJLFAMUZXA4RTSK74EAG/action/storage_attestation","attest_author":"https://pith.science/pith/GXPPOUBJLFAMUZXA4RTSK74EAG/action/author_attestation","sign_citation":"https://pith.science/pith/GXPPOUBJLFAMUZXA4RTSK74EAG/action/citation_signature","submit_replication":"https://pith.science/pith/GXPPOUBJLFAMUZXA4RTSK74EAG/action/replication_record"}},"created_at":"2026-07-05T06:19:54.711481+00:00","updated_at":"2026-07-05T06:19:54.711481+00:00"}