{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:UVZKDAHOXZRQ6DIJ4GYMYQU2FA","short_pith_number":"pith:UVZKDAHO","schema_version":"1.0","canonical_sha256":"a572a180eebe630f0d09e1b0cc429a283d324363f09492459bb6e1b0308d570f","source":{"kind":"arxiv","id":"2601.18778","version":3},"attestation_state":"computed","paper":{"title":"Teaching Models to Teach Themselves: Reasoning at the Edge of Learnability","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Ariel Kwiatkowski, John Quan, Julia Kempe, Kartik Ahuja, Shobhita Sundaram, Yann Ollivier","submitted_at":"2026-01-26T18:46:56Z","abstract_excerpt":"RL methods for scaling large reasoning models stall on datasets with low initial success rates, and thus little training signal. We investigate a fundamental question: Can a pretrained LLM leverage latent knowledge to generate an automated curriculum for problems it cannot solve? We explore this with SOAR: An asymmetric self-play framework that uses meta-RL to surface these pedagogical signals. A teacher model proposes synthetic problems for a student model, and is rewarded with its improvement on a subset of hard problems, thus grounding the curriculum in real student progress rather than int"},"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":"2601.18778","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-01-26T18:46:56Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"483cc36cb366f65344686c123a026b4468752615d4ccfdd4875583dbd8d6e703","abstract_canon_sha256":"c08f909e3eb7ba9d4518f823275984ff4aa05513514306734a65cf56cffdf58d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-01T01:18:24.162343Z","signature_b64":"p7NDKaiJ0bnzpHuZTy+K9jvoNgAJh/o9/ayWmOrnEz0SOCHJxb8fKFe0DztLtbaUG+//3gWdWF85ge7FC/Q1Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a572a180eebe630f0d09e1b0cc429a283d324363f09492459bb6e1b0308d570f","last_reissued_at":"2026-07-01T01:18:24.161911Z","signature_status":"signed_v1","first_computed_at":"2026-07-01T01:18:24.161911Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Teaching Models to Teach Themselves: Reasoning at the Edge of Learnability","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Ariel Kwiatkowski, John Quan, Julia Kempe, Kartik Ahuja, Shobhita Sundaram, Yann Ollivier","submitted_at":"2026-01-26T18:46:56Z","abstract_excerpt":"RL methods for scaling large reasoning models stall on datasets with low initial success rates, and thus little training signal. We investigate a fundamental question: Can a pretrained LLM leverage latent knowledge to generate an automated curriculum for problems it cannot solve? We explore this with SOAR: An asymmetric self-play framework that uses meta-RL to surface these pedagogical signals. A teacher model proposes synthetic problems for a student model, and is rewarded with its improvement on a subset of hard problems, thus grounding the curriculum in real student progress rather than int"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.18778","kind":"arxiv","version":3},"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/2601.18778/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":"2601.18778","created_at":"2026-07-01T01:18:24.161973+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.18778v3","created_at":"2026-07-01T01:18:24.161973+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.18778","created_at":"2026-07-01T01:18:24.161973+00:00"},{"alias_kind":"pith_short_12","alias_value":"UVZKDAHOXZRQ","created_at":"2026-07-01T01:18:24.161973+00:00"},{"alias_kind":"pith_short_16","alias_value":"UVZKDAHOXZRQ6DIJ","created_at":"2026-07-01T01:18:24.161973+00:00"},{"alias_kind":"pith_short_8","alias_value":"UVZKDAHO","created_at":"2026-07-01T01:18:24.161973+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2605.16727","citing_title":"PopuLoRA: Co-Evolving LLM Populations for Reasoning Self-Play","ref_index":48,"is_internal_anchor":true},{"citing_arxiv_id":"2604.00860","citing_title":"Policy Improvement Reinforcement Learning","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"2604.20209","citing_title":"Scaling Self-Play with Self-Guidance","ref_index":35,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UVZKDAHOXZRQ6DIJ4GYMYQU2FA","json":"https://pith.science/pith/UVZKDAHOXZRQ6DIJ4GYMYQU2FA.json","graph_json":"https://pith.science/api/pith-number/UVZKDAHOXZRQ6DIJ4GYMYQU2FA/graph.json","events_json":"https://pith.science/api/pith-number/UVZKDAHOXZRQ6DIJ4GYMYQU2FA/events.json","paper":"https://pith.science/paper/UVZKDAHO"},"agent_actions":{"view_html":"https://pith.science/pith/UVZKDAHOXZRQ6DIJ4GYMYQU2FA","download_json":"https://pith.science/pith/UVZKDAHOXZRQ6DIJ4GYMYQU2FA.json","view_paper":"https://pith.science/paper/UVZKDAHO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.18778&json=true","fetch_graph":"https://pith.science/api/pith-number/UVZKDAHOXZRQ6DIJ4GYMYQU2FA/graph.json","fetch_events":"https://pith.science/api/pith-number/UVZKDAHOXZRQ6DIJ4GYMYQU2FA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UVZKDAHOXZRQ6DIJ4GYMYQU2FA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UVZKDAHOXZRQ6DIJ4GYMYQU2FA/action/storage_attestation","attest_author":"https://pith.science/pith/UVZKDAHOXZRQ6DIJ4GYMYQU2FA/action/author_attestation","sign_citation":"https://pith.science/pith/UVZKDAHOXZRQ6DIJ4GYMYQU2FA/action/citation_signature","submit_replication":"https://pith.science/pith/UVZKDAHOXZRQ6DIJ4GYMYQU2FA/action/replication_record"}},"created_at":"2026-07-01T01:18:24.161973+00:00","updated_at":"2026-07-01T01:18:24.161973+00:00"}