{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:EASDRUWRLY6GKBGB7TJLWW5UQY","short_pith_number":"pith:EASDRUWR","schema_version":"1.0","canonical_sha256":"202438d2d15e3c6504c1fcd2bb5bb4860e7cbd039d5d85ab3187bf3f73e39da7","source":{"kind":"arxiv","id":"2511.00602","version":3},"attestation_state":"computed","paper":{"title":"OpenSIR: Open-Ended Self-Improving Reasoner","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aryo Pradipta Gema, Joshua Ong Jun Leang, Pasquale Minervini, Wai-Chung Kwan","submitted_at":"2025-11-01T16:08:28Z","abstract_excerpt":"Recent advances in large language model (LLM) reasoning through reinforcement learning rely on annotated datasets for verifiable rewards, which may limit models' ability to surpass human-level performance. While self-play offers a promising alternative, prior methods yield only marginal or even negative gains on post-trained models because they generate problems that cluster around familiar concepts rather than discovering novel ones. We introduce Open-Ended Self-Improving Reasoner (OpenSIR), a self-play framework in which a single LLM alternates teacher and student roles to generate and solve"},"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":"2511.00602","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-11-01T16:08:28Z","cross_cats_sorted":[],"title_canon_sha256":"5c437a889838cbaa7b958bcd1e30e236ee944cccde6363c7aa88a8ff874eb04c","abstract_canon_sha256":"3c9918d0fcff430bbb3b79504070edb444b61f2befb3c1ceab18d51ce88a7d4b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T01:12:48.047428Z","signature_b64":"vGrFLcjmtcq3GC+itI04N6IazN2k724npwh5sejdJXlSMBXWHa/w59wM0QsEOLIufzOCz0B6Ut1K/5bwCsdCBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"202438d2d15e3c6504c1fcd2bb5bb4860e7cbd039d5d85ab3187bf3f73e39da7","last_reissued_at":"2026-06-23T01:12:48.046827Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T01:12:48.046827Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"OpenSIR: Open-Ended Self-Improving Reasoner","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aryo Pradipta Gema, Joshua Ong Jun Leang, Pasquale Minervini, Wai-Chung Kwan","submitted_at":"2025-11-01T16:08:28Z","abstract_excerpt":"Recent advances in large language model (LLM) reasoning through reinforcement learning rely on annotated datasets for verifiable rewards, which may limit models' ability to surpass human-level performance. While self-play offers a promising alternative, prior methods yield only marginal or even negative gains on post-trained models because they generate problems that cluster around familiar concepts rather than discovering novel ones. We introduce Open-Ended Self-Improving Reasoner (OpenSIR), a self-play framework in which a single LLM alternates teacher and student roles to generate and solve"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2511.00602","kind":"arxiv","version":3},"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/2511.00602/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":"2511.00602","created_at":"2026-06-23T01:12:48.046903+00:00"},{"alias_kind":"arxiv_version","alias_value":"2511.00602v3","created_at":"2026-06-23T01:12:48.046903+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.00602","created_at":"2026-06-23T01:12:48.046903+00:00"},{"alias_kind":"pith_short_12","alias_value":"EASDRUWRLY6G","created_at":"2026-06-23T01:12:48.046903+00:00"},{"alias_kind":"pith_short_16","alias_value":"EASDRUWRLY6GKBGB","created_at":"2026-06-23T01:12:48.046903+00:00"},{"alias_kind":"pith_short_8","alias_value":"EASDRUWR","created_at":"2026-06-23T01:12:48.046903+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.16727","citing_title":"PopuLoRA: Co-Evolving LLM Populations for Reasoning Self-Play","ref_index":62,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EASDRUWRLY6GKBGB7TJLWW5UQY","json":"https://pith.science/pith/EASDRUWRLY6GKBGB7TJLWW5UQY.json","graph_json":"https://pith.science/api/pith-number/EASDRUWRLY6GKBGB7TJLWW5UQY/graph.json","events_json":"https://pith.science/api/pith-number/EASDRUWRLY6GKBGB7TJLWW5UQY/events.json","paper":"https://pith.science/paper/EASDRUWR"},"agent_actions":{"view_html":"https://pith.science/pith/EASDRUWRLY6GKBGB7TJLWW5UQY","download_json":"https://pith.science/pith/EASDRUWRLY6GKBGB7TJLWW5UQY.json","view_paper":"https://pith.science/paper/EASDRUWR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2511.00602&json=true","fetch_graph":"https://pith.science/api/pith-number/EASDRUWRLY6GKBGB7TJLWW5UQY/graph.json","fetch_events":"https://pith.science/api/pith-number/EASDRUWRLY6GKBGB7TJLWW5UQY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EASDRUWRLY6GKBGB7TJLWW5UQY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EASDRUWRLY6GKBGB7TJLWW5UQY/action/storage_attestation","attest_author":"https://pith.science/pith/EASDRUWRLY6GKBGB7TJLWW5UQY/action/author_attestation","sign_citation":"https://pith.science/pith/EASDRUWRLY6GKBGB7TJLWW5UQY/action/citation_signature","submit_replication":"https://pith.science/pith/EASDRUWRLY6GKBGB7TJLWW5UQY/action/replication_record"}},"created_at":"2026-06-23T01:12:48.046903+00:00","updated_at":"2026-06-23T01:12:48.046903+00:00"}