{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:QUZWYNORLJIQMM7OGIXDYZ67A3","short_pith_number":"pith:QUZWYNOR","schema_version":"1.0","canonical_sha256":"85336c35d15a510633ee322e3c67df06d730c451340228941d4f390d32bf8796","source":{"kind":"arxiv","id":"2606.03800","version":1},"attestation_state":"computed","paper":{"title":"Trading Human Curation for Synthetic Augmentation in RLVR","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Akshansh <last>, Leonardo Rosa Rodrigues, Mark E. Whiting, Michael Korostelev, Youssef Hassan","submitted_at":"2026-06-02T15:48:28Z","abstract_excerpt":"The supply of high-quality training tasks is a central bottleneck for reinforcement learning from verifiable rewards (RLVR) on agentic language models. Each task requires a sandboxed setup, a prompt, and a hand-authored reward function, and only tasks that pass a quality bar produce useful training signal. Hand-curation at this quality bar does not scale economically to the task counts effective RL training requires, and the substitution rate between automatically generated task variants and human-authored ones is not yet established. We investigate using pre-specified, gate-filtered augmentat"},"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":"2606.03800","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-02T15:48:28Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ccdfd49330a259b2cb7e89646ae4180f90aa3dc7415ce5454b7ad25eb282e153","abstract_canon_sha256":"a6d3601289ae3f0b6587df18435cb3f3dad96e3287272c3df668d29e550e4877"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-03T02:06:02.782086Z","signature_b64":"749sW3vkdblVMcUn+dJ8dVU0kak7YgYNvLP6CplMSA9WWuUOC06ABocMlbwlR0DOkNCq6t5QtonUj3F9N9BMAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"85336c35d15a510633ee322e3c67df06d730c451340228941d4f390d32bf8796","last_reissued_at":"2026-06-03T02:06:02.781635Z","signature_status":"signed_v1","first_computed_at":"2026-06-03T02:06:02.781635Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Trading Human Curation for Synthetic Augmentation in RLVR","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Akshansh <last>, Leonardo Rosa Rodrigues, Mark E. Whiting, Michael Korostelev, Youssef Hassan","submitted_at":"2026-06-02T15:48:28Z","abstract_excerpt":"The supply of high-quality training tasks is a central bottleneck for reinforcement learning from verifiable rewards (RLVR) on agentic language models. Each task requires a sandboxed setup, a prompt, and a hand-authored reward function, and only tasks that pass a quality bar produce useful training signal. Hand-curation at this quality bar does not scale economically to the task counts effective RL training requires, and the substitution rate between automatically generated task variants and human-authored ones is not yet established. We investigate using pre-specified, gate-filtered augmentat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.03800","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/2606.03800/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":"2606.03800","created_at":"2026-06-03T02:06:02.781695+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.03800v1","created_at":"2026-06-03T02:06:02.781695+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.03800","created_at":"2026-06-03T02:06:02.781695+00:00"},{"alias_kind":"pith_short_12","alias_value":"QUZWYNORLJIQ","created_at":"2026-06-03T02:06:02.781695+00:00"},{"alias_kind":"pith_short_16","alias_value":"QUZWYNORLJIQMM7O","created_at":"2026-06-03T02:06:02.781695+00:00"},{"alias_kind":"pith_short_8","alias_value":"QUZWYNOR","created_at":"2026-06-03T02:06:02.781695+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/QUZWYNORLJIQMM7OGIXDYZ67A3","json":"https://pith.science/pith/QUZWYNORLJIQMM7OGIXDYZ67A3.json","graph_json":"https://pith.science/api/pith-number/QUZWYNORLJIQMM7OGIXDYZ67A3/graph.json","events_json":"https://pith.science/api/pith-number/QUZWYNORLJIQMM7OGIXDYZ67A3/events.json","paper":"https://pith.science/paper/QUZWYNOR"},"agent_actions":{"view_html":"https://pith.science/pith/QUZWYNORLJIQMM7OGIXDYZ67A3","download_json":"https://pith.science/pith/QUZWYNORLJIQMM7OGIXDYZ67A3.json","view_paper":"https://pith.science/paper/QUZWYNOR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.03800&json=true","fetch_graph":"https://pith.science/api/pith-number/QUZWYNORLJIQMM7OGIXDYZ67A3/graph.json","fetch_events":"https://pith.science/api/pith-number/QUZWYNORLJIQMM7OGIXDYZ67A3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QUZWYNORLJIQMM7OGIXDYZ67A3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QUZWYNORLJIQMM7OGIXDYZ67A3/action/storage_attestation","attest_author":"https://pith.science/pith/QUZWYNORLJIQMM7OGIXDYZ67A3/action/author_attestation","sign_citation":"https://pith.science/pith/QUZWYNORLJIQMM7OGIXDYZ67A3/action/citation_signature","submit_replication":"https://pith.science/pith/QUZWYNORLJIQMM7OGIXDYZ67A3/action/replication_record"}},"created_at":"2026-06-03T02:06:02.781695+00:00","updated_at":"2026-06-03T02:06:02.781695+00:00"}