{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:V4RDKFGP6YR2PIL76BEYQSBT7J","short_pith_number":"pith:V4RDKFGP","schema_version":"1.0","canonical_sha256":"af223514cff623a7a17ff049884833fa7e08da7decc0a91b1499a991e46d2ee2","source":{"kind":"arxiv","id":"2501.09804","version":1},"attestation_state":"computed","paper":{"title":"Enhancing Generalization in Chain of Thought Reasoning for Smaller Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Boyu Wang, Charles Ling, Dingyi Jiang, Maxwell J. Yin, Yongbing Chen","submitted_at":"2025-01-16T19:23:11Z","abstract_excerpt":"Chain-of-Thought (CoT) reasoning in smaller language models is a challenging natural language process problem yet highly desirable in many real-life applications. Existing CoT knowledge distillation methods often suffer from overly conservative memorization in smaller LLMs, leading to low generalization confidence. As fully preserving the CoT ability of teacher model is impossible, we hypothesize that adversarial CoT fine-tuning is crucial for developing smaller LLM with robust CoT generalization. To this end, we propose \\textit{PRompt-Assisted Domain-Adversarial fine-tuning} (PRADA), a princi"},"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":"2501.09804","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-01-16T19:23:11Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"8a3ec47993c96f3f72baf956b4528db2562bfe23d09c04bd5daf71c90480ba44","abstract_canon_sha256":"3e4aa38a74feca5a7e3348f133e73f878fa92858bcb1198f3221d921347745c7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:02:11.404780Z","signature_b64":"+5X9J6xhxKBsys1uNHs+s4XbflvTE1tifgIWbSN6/awF6WF/4OULDBp2lWMyxcSNCA23GaOVnTdEltfjj7sGBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"af223514cff623a7a17ff049884833fa7e08da7decc0a91b1499a991e46d2ee2","last_reissued_at":"2026-07-05T10:02:11.404358Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:02:11.404358Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Enhancing Generalization in Chain of Thought Reasoning for Smaller Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Boyu Wang, Charles Ling, Dingyi Jiang, Maxwell J. Yin, Yongbing Chen","submitted_at":"2025-01-16T19:23:11Z","abstract_excerpt":"Chain-of-Thought (CoT) reasoning in smaller language models is a challenging natural language process problem yet highly desirable in many real-life applications. Existing CoT knowledge distillation methods often suffer from overly conservative memorization in smaller LLMs, leading to low generalization confidence. As fully preserving the CoT ability of teacher model is impossible, we hypothesize that adversarial CoT fine-tuning is crucial for developing smaller LLM with robust CoT generalization. To this end, we propose \\textit{PRompt-Assisted Domain-Adversarial fine-tuning} (PRADA), a princi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.09804","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/2501.09804/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":"2501.09804","created_at":"2026-07-05T10:02:11.404415+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.09804v1","created_at":"2026-07-05T10:02:11.404415+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.09804","created_at":"2026-07-05T10:02:11.404415+00:00"},{"alias_kind":"pith_short_12","alias_value":"V4RDKFGP6YR2","created_at":"2026-07-05T10:02:11.404415+00:00"},{"alias_kind":"pith_short_16","alias_value":"V4RDKFGP6YR2PIL7","created_at":"2026-07-05T10:02:11.404415+00:00"},{"alias_kind":"pith_short_8","alias_value":"V4RDKFGP","created_at":"2026-07-05T10:02:11.404415+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01168","citing_title":"Thinking Economically: A Hierarchical Framework for Adaptive-Complexity Reasoning in LLMs","ref_index":40,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/V4RDKFGP6YR2PIL76BEYQSBT7J","json":"https://pith.science/pith/V4RDKFGP6YR2PIL76BEYQSBT7J.json","graph_json":"https://pith.science/api/pith-number/V4RDKFGP6YR2PIL76BEYQSBT7J/graph.json","events_json":"https://pith.science/api/pith-number/V4RDKFGP6YR2PIL76BEYQSBT7J/events.json","paper":"https://pith.science/paper/V4RDKFGP"},"agent_actions":{"view_html":"https://pith.science/pith/V4RDKFGP6YR2PIL76BEYQSBT7J","download_json":"https://pith.science/pith/V4RDKFGP6YR2PIL76BEYQSBT7J.json","view_paper":"https://pith.science/paper/V4RDKFGP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.09804&json=true","fetch_graph":"https://pith.science/api/pith-number/V4RDKFGP6YR2PIL76BEYQSBT7J/graph.json","fetch_events":"https://pith.science/api/pith-number/V4RDKFGP6YR2PIL76BEYQSBT7J/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V4RDKFGP6YR2PIL76BEYQSBT7J/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V4RDKFGP6YR2PIL76BEYQSBT7J/action/storage_attestation","attest_author":"https://pith.science/pith/V4RDKFGP6YR2PIL76BEYQSBT7J/action/author_attestation","sign_citation":"https://pith.science/pith/V4RDKFGP6YR2PIL76BEYQSBT7J/action/citation_signature","submit_replication":"https://pith.science/pith/V4RDKFGP6YR2PIL76BEYQSBT7J/action/replication_record"}},"created_at":"2026-07-05T10:02:11.404415+00:00","updated_at":"2026-07-05T10:02:11.404415+00:00"}