{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:C7Y3M7G4XKCWKFMLDGOISJJKNX","short_pith_number":"pith:C7Y3M7G4","schema_version":"1.0","canonical_sha256":"17f1b67cdcba8565158b199c89252a6df2339321e26e07bde1637fe1b0acbbd4","source":{"kind":"arxiv","id":"2605.01248","version":3},"attestation_state":"computed","paper":{"title":"$S^3$-R1: Learning to Retrieve and Answer Step-by-Step with Synthetic Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Coupling synthetic multi-hop questions with rewards for search steps and answers enables models to learn effective retrieval strategies and generalize up to 10% better out of domain.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Akhil Udathu, Atharva Parulekar, Harsh Goel, Pradnesh Kalkar, Susmija Jabbireddy","submitted_at":"2026-05-02T05:01:05Z","abstract_excerpt":"Reinforcement learning (RL) post-training has enabled newer capabilities in models, such as agentic tool-use for search. However, these models struggle primarily due to limitations with sparse outcome-based rewards and a lack of training data that encapsulates questions of differing hardness, which results in models not performing deeper searches with tools to collect evidence for question-answering. To address these limitations, we introduce S^3-R1 (Synthetic data and stabilized Search R1), a framework that couples a data-centric approach with denser learning signals. We first develop a synth"},"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":"2605.01248","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-02T05:01:05Z","cross_cats_sorted":[],"title_canon_sha256":"1f01da6397752e8cb232c781c825d0d88c42bc368335cebbbc1daf20146ac842","abstract_canon_sha256":"6f4f561a01ded9716eeeb40f87ff4b3fb14a5d398fe514248c57dd4a58fd4c93"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-10T01:08:36.074547Z","signature_b64":"r2Io5K5J+/Jk0hOCkjBE0rmWLcxbVLODEY9w1HSfKBSVuWImrNK6DlameTCIfUh5dS5W/2AJW81Fs2xduQ1rBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"17f1b67cdcba8565158b199c89252a6df2339321e26e07bde1637fe1b0acbbd4","last_reissued_at":"2026-06-10T01:08:36.073608Z","signature_status":"signed_v1","first_computed_at":"2026-06-10T01:08:36.073608Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"$S^3$-R1: Learning to Retrieve and Answer Step-by-Step with Synthetic Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Coupling synthetic multi-hop questions with rewards for search steps and answers enables models to learn effective retrieval strategies and generalize up to 10% better out of domain.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Akhil Udathu, Atharva Parulekar, Harsh Goel, Pradnesh Kalkar, Susmija Jabbireddy","submitted_at":"2026-05-02T05:01:05Z","abstract_excerpt":"Reinforcement learning (RL) post-training has enabled newer capabilities in models, such as agentic tool-use for search. However, these models struggle primarily due to limitations with sparse outcome-based rewards and a lack of training data that encapsulates questions of differing hardness, which results in models not performing deeper searches with tools to collect evidence for question-answering. To address these limitations, we introduce S^3-R1 (Synthetic data and stabilized Search R1), a framework that couples a data-centric approach with denser learning signals. We first develop a synth"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our evaluations show that S^3-R1 outperforms existing baselines by learning more effective search and synthesis strategies, yielding up to a 10% improvement in robust generalization on out-of-domain datasets.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The synthetic generation and retrieval-based verification pipeline produces questions of genuinely intermediate difficulty that transfer to real user queries without introducing distribution shift or annotation artifacts that inflate measured gains.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"S^3-R1 generates synthetic intermediate-difficulty multi-hop questions and applies dense rewards for search quality plus answer correctness, yielding up to 10% better out-of-domain generalization than baselines.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Coupling synthetic multi-hop questions with rewards for search steps and answers enables models to learn effective retrieval strategies and generalize up to 10% better out of domain.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"1e0be816271a93b5b2ce8a405475b9b96d5fd22abc48d355f12fbd768549b879"},"source":{"id":"2605.01248","kind":"arxiv","version":3},"verdict":{"id":"5a42145d-1340-44a4-9004-dae1b835fed5","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-09T15:05:21.542628Z","strongest_claim":"Our evaluations show that S^3-R1 outperforms existing baselines by learning more effective search and synthesis strategies, yielding up to a 10% improvement in robust generalization on out-of-domain datasets.","one_line_summary":"S^3-R1 generates synthetic intermediate-difficulty multi-hop questions and applies dense rewards for search quality plus answer correctness, yielding up to 10% better out-of-domain generalization than baselines.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The synthetic generation and retrieval-based verification pipeline produces questions of genuinely intermediate difficulty that transfer to real user queries without introducing distribution shift or annotation artifacts that inflate measured gains.","pith_extraction_headline":"Coupling synthetic multi-hop questions with rewards for search steps and answers enables models to learn effective retrieval strategies and generalize up to 10% better out of domain."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.01248/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-20T18:36:11.355256Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T17:28:38.166668Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"304b5c8b06090ffc52ef51d631cf34569b03302993c92030bfeae2c01d150f4b"},"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":"2605.01248","created_at":"2026-06-10T01:08:36.073739+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.01248v3","created_at":"2026-06-10T01:08:36.073739+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.01248","created_at":"2026-06-10T01:08:36.073739+00:00"},{"alias_kind":"pith_short_12","alias_value":"C7Y3M7G4XKCW","created_at":"2026-06-10T01:08:36.073739+00:00"},{"alias_kind":"pith_short_16","alias_value":"C7Y3M7G4XKCWKFML","created_at":"2026-06-10T01:08:36.073739+00:00"},{"alias_kind":"pith_short_8","alias_value":"C7Y3M7G4","created_at":"2026-06-10T01:08:36.073739+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/C7Y3M7G4XKCWKFMLDGOISJJKNX","json":"https://pith.science/pith/C7Y3M7G4XKCWKFMLDGOISJJKNX.json","graph_json":"https://pith.science/api/pith-number/C7Y3M7G4XKCWKFMLDGOISJJKNX/graph.json","events_json":"https://pith.science/api/pith-number/C7Y3M7G4XKCWKFMLDGOISJJKNX/events.json","paper":"https://pith.science/paper/C7Y3M7G4"},"agent_actions":{"view_html":"https://pith.science/pith/C7Y3M7G4XKCWKFMLDGOISJJKNX","download_json":"https://pith.science/pith/C7Y3M7G4XKCWKFMLDGOISJJKNX.json","view_paper":"https://pith.science/paper/C7Y3M7G4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.01248&json=true","fetch_graph":"https://pith.science/api/pith-number/C7Y3M7G4XKCWKFMLDGOISJJKNX/graph.json","fetch_events":"https://pith.science/api/pith-number/C7Y3M7G4XKCWKFMLDGOISJJKNX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/C7Y3M7G4XKCWKFMLDGOISJJKNX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/C7Y3M7G4XKCWKFMLDGOISJJKNX/action/storage_attestation","attest_author":"https://pith.science/pith/C7Y3M7G4XKCWKFMLDGOISJJKNX/action/author_attestation","sign_citation":"https://pith.science/pith/C7Y3M7G4XKCWKFMLDGOISJJKNX/action/citation_signature","submit_replication":"https://pith.science/pith/C7Y3M7G4XKCWKFMLDGOISJJKNX/action/replication_record"}},"created_at":"2026-06-10T01:08:36.073739+00:00","updated_at":"2026-06-10T01:08:36.073739+00:00"}