{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GIDL7EDIPK547X4WFQF343VGDI","short_pith_number":"pith:GIDL7EDI","schema_version":"1.0","canonical_sha256":"3206bf90687abbcfdf962c0bbe6ea61a259ffc8a711e099e8cd6a8c01927ea9e","source":{"kind":"arxiv","id":"2605.09270","version":2},"attestation_state":"computed","paper":{"title":"Memorize Theorems, Not Instances: Probing SFT Generalization through Mathematical Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Supervised fine-tuning for math reasoning succeeds when models learn to apply theorems explicitly instead of memorizing individual problem-answer pairs.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jing Lei, Mengyu Yang, Ruiying Peng, Xiaohui Li, Xinlei Chen, Xueyu Wu","submitted_at":"2026-05-10T02:39:05Z","abstract_excerpt":"Supervised Fine-Tuning (SFT) is widely used for task-specific adaptation, yet recent work shows it systematically undermines reasoning generalization. We argue the root cause is not memorization itself, but its target: vanilla SFT drives models to exploit and memorize spurious surface correlations in problem-solution pairs, leaving them brittle to superficial input variations. To address this, we propose Theorem-SFT, which reorients supervision toward explicit theorem application by teaching models how rules are invoked rather than what answers look like. Theorem-SFT yields consistent gains ac"},"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":true},"canonical_record":{"source":{"id":"2605.09270","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-10T02:39:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2157ed097ea7cbccee21b89c015f0df8544df3e69a5bc8306a22aa8959162c6d","abstract_canon_sha256":"881f5c372f9beeea9741eb229f4420493808e42c9867d83c4256600ac50ab2cc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T02:04:12.659552Z","signature_b64":"i24naM5FCvKmx0u21Rv+fuhGJB7oRATMLbjtHsCOBm4QtVwdPPWznmonmuYdSRcTg7akYsZqbIKpmp+Eht2aDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3206bf90687abbcfdf962c0bbe6ea61a259ffc8a711e099e8cd6a8c01927ea9e","last_reissued_at":"2026-05-26T02:04:12.658778Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T02:04:12.658778Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Memorize Theorems, Not Instances: Probing SFT Generalization through Mathematical Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Supervised fine-tuning for math reasoning succeeds when models learn to apply theorems explicitly instead of memorizing individual problem-answer pairs.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jing Lei, Mengyu Yang, Ruiying Peng, Xiaohui Li, Xinlei Chen, Xueyu Wu","submitted_at":"2026-05-10T02:39:05Z","abstract_excerpt":"Supervised Fine-Tuning (SFT) is widely used for task-specific adaptation, yet recent work shows it systematically undermines reasoning generalization. We argue the root cause is not memorization itself, but its target: vanilla SFT drives models to exploit and memorize spurious surface correlations in problem-solution pairs, leaving them brittle to superficial input variations. To address this, we propose Theorem-SFT, which reorients supervision toward explicit theorem application by teaching models how rules are invoked rather than what answers look like. Theorem-SFT yields consistent gains ac"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Generalization failures stem not from memorization as a mechanism, but from memorizing the wrong inductive targets.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the reported performance gains are caused by the shift to theorem-level supervision rather than by other unspecified differences in data construction, prompting, or training hyperparameters between vanilla SFT and Theorem-SFT.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Theorem-SFT improves mathematical reasoning generalization by teaching theorem application rather than instance memorization, delivering gains of +8.8% on MATH and +20.27% on GeoQA across model families.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Supervised fine-tuning for math reasoning succeeds when models learn to apply theorems explicitly instead of memorizing individual problem-answer pairs.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"2ac3c3605335d47feb2fce4ee18293ae8490f872b3ee468d2f940c37bf81d2ad"},"source":{"id":"2605.09270","kind":"arxiv","version":2},"verdict":{"id":"d59a00fe-0325-4e05-a41f-c3b14d40b8ff","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-12T04:47:26.502142Z","strongest_claim":"Generalization failures stem not from memorization as a mechanism, but from memorizing the wrong inductive targets.","one_line_summary":"Theorem-SFT improves mathematical reasoning generalization by teaching theorem application rather than instance memorization, delivering gains of +8.8% on MATH and +20.27% on GeoQA across model families.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the reported performance gains are caused by the shift to theorem-level supervision rather than by other unspecified differences in data construction, prompting, or training hyperparameters between vanilla SFT and Theorem-SFT.","pith_extraction_headline":"Supervised fine-tuning for math reasoning succeeds when models learn to apply theorems explicitly instead of memorizing individual problem-answer pairs."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.09270/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T08:02:08.944337Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T20:34:31.582370Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T13:31:17.837425Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T10:22:17.241840Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"83df298cf3e2c2c390afdf01fcaf3a6efeeb248bd79cbd9c70a7c5a47a0dae34"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"e90073efb785a850c8bbcb483d17bd3ecc03cc98414f435b4534a6c1a00a2f58"},"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":"2605.09270","created_at":"2026-05-26T02:04:12.658876+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.09270v2","created_at":"2026-05-26T02:04:12.658876+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.09270","created_at":"2026-05-26T02:04:12.658876+00:00"},{"alias_kind":"pith_short_12","alias_value":"GIDL7EDIPK54","created_at":"2026-05-26T02:04:12.658876+00:00"},{"alias_kind":"pith_short_16","alias_value":"GIDL7EDIPK547X4W","created_at":"2026-05-26T02:04:12.658876+00:00"},{"alias_kind":"pith_short_8","alias_value":"GIDL7EDI","created_at":"2026-05-26T02:04:12.658876+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GIDL7EDIPK547X4WFQF343VGDI","json":"https://pith.science/pith/GIDL7EDIPK547X4WFQF343VGDI.json","graph_json":"https://pith.science/api/pith-number/GIDL7EDIPK547X4WFQF343VGDI/graph.json","events_json":"https://pith.science/api/pith-number/GIDL7EDIPK547X4WFQF343VGDI/events.json","paper":"https://pith.science/paper/GIDL7EDI"},"agent_actions":{"view_html":"https://pith.science/pith/GIDL7EDIPK547X4WFQF343VGDI","download_json":"https://pith.science/pith/GIDL7EDIPK547X4WFQF343VGDI.json","view_paper":"https://pith.science/paper/GIDL7EDI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.09270&json=true","fetch_graph":"https://pith.science/api/pith-number/GIDL7EDIPK547X4WFQF343VGDI/graph.json","fetch_events":"https://pith.science/api/pith-number/GIDL7EDIPK547X4WFQF343VGDI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GIDL7EDIPK547X4WFQF343VGDI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GIDL7EDIPK547X4WFQF343VGDI/action/storage_attestation","attest_author":"https://pith.science/pith/GIDL7EDIPK547X4WFQF343VGDI/action/author_attestation","sign_citation":"https://pith.science/pith/GIDL7EDIPK547X4WFQF343VGDI/action/citation_signature","submit_replication":"https://pith.science/pith/GIDL7EDIPK547X4WFQF343VGDI/action/replication_record"}},"created_at":"2026-05-26T02:04:12.658876+00:00","updated_at":"2026-05-26T02:04:12.658876+00:00"}