{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:7QGPUZIQL655MA4JKD53C2HTL2","short_pith_number":"pith:7QGPUZIQ","schema_version":"1.0","canonical_sha256":"fc0cfa65105fbbd6038950fbb168f35eaff76776116c50c4d573e67b20435491","source":{"kind":"arxiv","id":"2606.13680","version":1},"attestation_state":"computed","paper":{"title":"Learning to Reason by Analogy via Retrieval-Augmented Reinforcement Fine-Tuning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Avinash Atreya, Chun-cheng Jason Chen, Hanjie Chen, Qi Ma, Vicente Ordonez, Xintao Chen, Zilin Xiao","submitted_at":"2026-06-11T17:59:52Z","abstract_excerpt":"Retrieval-augmented generation (RAG) has become a standard mechanism for grounding language models in external knowledge, yet conventional retrieval based on lexical or semantic similarity is poorly suited for complex reasoning tasks: a semantically similar problem may demand an entirely different solution strategy, while a superficially different problem may share the same underlying reasoning pattern. We propose Retrieval-Augmented Reinforcement Fine-Tuning (RA-RFT), a post-training framework that teaches language models to reason by analogy. RA-RFT uses gold-relevance distillation to train "},"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":"2606.13680","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-11T17:59:52Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"944bb0857707ae4dd0d0fc8a8be196ea6380aad69c911e33066a1c17f18f63a9","abstract_canon_sha256":"522ee4f559598ce40637fd405b808ede5334b8ca5e3793271d700297e5853ffe"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-12T01:10:23.195363Z","signature_b64":"q2PdM7N69MY74kuUz4ZDm2CljRE8Yo0Ov9zb3zVTG6L6QuohLIXcGxyYGPMRuAdjZ+Hubr1kv9i04RH7+WBqBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fc0cfa65105fbbd6038950fbb168f35eaff76776116c50c4d573e67b20435491","last_reissued_at":"2026-06-12T01:10:23.194535Z","signature_status":"signed_v1","first_computed_at":"2026-06-12T01:10:23.194535Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning to Reason by Analogy via Retrieval-Augmented Reinforcement Fine-Tuning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Avinash Atreya, Chun-cheng Jason Chen, Hanjie Chen, Qi Ma, Vicente Ordonez, Xintao Chen, Zilin Xiao","submitted_at":"2026-06-11T17:59:52Z","abstract_excerpt":"Retrieval-augmented generation (RAG) has become a standard mechanism for grounding language models in external knowledge, yet conventional retrieval based on lexical or semantic similarity is poorly suited for complex reasoning tasks: a semantically similar problem may demand an entirely different solution strategy, while a superficially different problem may share the same underlying reasoning pattern. We propose Retrieval-Augmented Reinforcement Fine-Tuning (RA-RFT), a post-training framework that teaches language models to reason by analogy. RA-RFT uses gold-relevance distillation to train "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.13680","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/2606.13680/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":"2606.13680","created_at":"2026-06-12T01:10:23.194662+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.13680v1","created_at":"2026-06-12T01:10:23.194662+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.13680","created_at":"2026-06-12T01:10:23.194662+00:00"},{"alias_kind":"pith_short_12","alias_value":"7QGPUZIQL655","created_at":"2026-06-12T01:10:23.194662+00:00"},{"alias_kind":"pith_short_16","alias_value":"7QGPUZIQL655MA4J","created_at":"2026-06-12T01:10:23.194662+00:00"},{"alias_kind":"pith_short_8","alias_value":"7QGPUZIQ","created_at":"2026-06-12T01:10:23.194662+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7QGPUZIQL655MA4JKD53C2HTL2","json":"https://pith.science/pith/7QGPUZIQL655MA4JKD53C2HTL2.json","graph_json":"https://pith.science/api/pith-number/7QGPUZIQL655MA4JKD53C2HTL2/graph.json","events_json":"https://pith.science/api/pith-number/7QGPUZIQL655MA4JKD53C2HTL2/events.json","paper":"https://pith.science/paper/7QGPUZIQ"},"agent_actions":{"view_html":"https://pith.science/pith/7QGPUZIQL655MA4JKD53C2HTL2","download_json":"https://pith.science/pith/7QGPUZIQL655MA4JKD53C2HTL2.json","view_paper":"https://pith.science/paper/7QGPUZIQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.13680&json=true","fetch_graph":"https://pith.science/api/pith-number/7QGPUZIQL655MA4JKD53C2HTL2/graph.json","fetch_events":"https://pith.science/api/pith-number/7QGPUZIQL655MA4JKD53C2HTL2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7QGPUZIQL655MA4JKD53C2HTL2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7QGPUZIQL655MA4JKD53C2HTL2/action/storage_attestation","attest_author":"https://pith.science/pith/7QGPUZIQL655MA4JKD53C2HTL2/action/author_attestation","sign_citation":"https://pith.science/pith/7QGPUZIQL655MA4JKD53C2HTL2/action/citation_signature","submit_replication":"https://pith.science/pith/7QGPUZIQL655MA4JKD53C2HTL2/action/replication_record"}},"created_at":"2026-06-12T01:10:23.194662+00:00","updated_at":"2026-06-12T01:10:23.194662+00:00"}