{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:4H7LDMYF2HZSJ35LMFQ2KYRJDN","short_pith_number":"pith:4H7LDMYF","schema_version":"1.0","canonical_sha256":"e1feb1b305d1f324efab6161a562291b5c59ee1990d406d0e9075b555e9d5599","source":{"kind":"arxiv","id":"2505.21493","version":1},"attestation_state":"computed","paper":{"title":"Reinforcing General Reasoning without Verifiers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Anya Sims, Chao Du, Chongxuan Li, Haonan Wang, Liang Wang, Min Lin, Tianyu Pang, Xiangxin Zhou, Zichen Liu","submitted_at":"2025-05-27T17:56:27Z","abstract_excerpt":"The recent paradigm shift towards training large language models (LLMs) using DeepSeek-R1-Zero-style reinforcement learning (RL) on verifiable rewards has led to impressive advancements in code and mathematical reasoning. However, this methodology is limited to tasks where rule-based answer verification is possible and does not naturally extend to real-world domains such as chemistry, healthcare, engineering, law, biology, business, and economics. Current practical workarounds use an additional LLM as a model-based verifier; however, this introduces issues such as reliance on a strong verifier"},"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":"2505.21493","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-27T17:56:27Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"58c1b0e670a17182215c7c346655631d450a494274814fcd36231bd1e737466c","abstract_canon_sha256":"7ffc631c84742bc6c0104b99ae447e4145d4c693466fad332a6ef096e4a56279"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:10:42.750895Z","signature_b64":"NvyzoJleLxDwAb/+hCGi2TYPeoLP0YSCjuTd6Nu/NgWDqIEdphgFEKxUcCOxCj+jVcrtorjodOQu44GNYsUTDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e1feb1b305d1f324efab6161a562291b5c59ee1990d406d0e9075b555e9d5599","last_reissued_at":"2026-07-05T11:10:42.750401Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:10:42.750401Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reinforcing General Reasoning without Verifiers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Anya Sims, Chao Du, Chongxuan Li, Haonan Wang, Liang Wang, Min Lin, Tianyu Pang, Xiangxin Zhou, Zichen Liu","submitted_at":"2025-05-27T17:56:27Z","abstract_excerpt":"The recent paradigm shift towards training large language models (LLMs) using DeepSeek-R1-Zero-style reinforcement learning (RL) on verifiable rewards has led to impressive advancements in code and mathematical reasoning. However, this methodology is limited to tasks where rule-based answer verification is possible and does not naturally extend to real-world domains such as chemistry, healthcare, engineering, law, biology, business, and economics. Current practical workarounds use an additional LLM as a model-based verifier; however, this introduces issues such as reliance on a strong verifier"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.21493","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/2505.21493/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":"2505.21493","created_at":"2026-07-05T11:10:42.750464+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.21493v1","created_at":"2026-07-05T11:10:42.750464+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.21493","created_at":"2026-07-05T11:10:42.750464+00:00"},{"alias_kind":"pith_short_12","alias_value":"4H7LDMYF2HZS","created_at":"2026-07-05T11:10:42.750464+00:00"},{"alias_kind":"pith_short_16","alias_value":"4H7LDMYF2HZSJ35L","created_at":"2026-07-05T11:10:42.750464+00:00"},{"alias_kind":"pith_short_8","alias_value":"4H7LDMYF","created_at":"2026-07-05T11:10:42.750464+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":11,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01249","citing_title":"Trust Region On-Policy Distillation","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00609","citing_title":"CARE-RL: Capability-Aware Reinforcement Learning for Mitigating Cross-Domain Conflicts","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2602.12579","citing_title":"VI-CuRL: Stabilizing Verifier-Independent RL Reasoning via Confidence-Guided Variance Reduction","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2504.01990","citing_title":"Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems","ref_index":167,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10810","citing_title":"Likelihood scoring for continuations of mathematical text: a self-supervised benchmark with tests for shortcut vulnerabilities","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2506.13351","citing_title":"Direct Reasoning Optimization: Token-Level Reasoning Reflectivity Meets Rubric Gates for Unverifiable Tasks","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2509.14234","citing_title":"Compute as Teacher: Turning Inference Compute Into Reference-Free Supervision","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2511.09907","citing_title":"Learning to Pose Problems: Reasoning-Driven and Solver-Adaptive Data Synthesis","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09959","citing_title":"G-Zero: Self-Play for Open-Ended Generation from Zero Data","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02913","citing_title":"Generate, Filter, Control, Replay: A Comprehensive Survey of Rollout Strategies for LLM Reinforcement Learning","ref_index":187,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18493","citing_title":"Too Correct to Learn: Reinforcement Learning on Saturated Reasoning Data","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4H7LDMYF2HZSJ35LMFQ2KYRJDN","json":"https://pith.science/pith/4H7LDMYF2HZSJ35LMFQ2KYRJDN.json","graph_json":"https://pith.science/api/pith-number/4H7LDMYF2HZSJ35LMFQ2KYRJDN/graph.json","events_json":"https://pith.science/api/pith-number/4H7LDMYF2HZSJ35LMFQ2KYRJDN/events.json","paper":"https://pith.science/paper/4H7LDMYF"},"agent_actions":{"view_html":"https://pith.science/pith/4H7LDMYF2HZSJ35LMFQ2KYRJDN","download_json":"https://pith.science/pith/4H7LDMYF2HZSJ35LMFQ2KYRJDN.json","view_paper":"https://pith.science/paper/4H7LDMYF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.21493&json=true","fetch_graph":"https://pith.science/api/pith-number/4H7LDMYF2HZSJ35LMFQ2KYRJDN/graph.json","fetch_events":"https://pith.science/api/pith-number/4H7LDMYF2HZSJ35LMFQ2KYRJDN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4H7LDMYF2HZSJ35LMFQ2KYRJDN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4H7LDMYF2HZSJ35LMFQ2KYRJDN/action/storage_attestation","attest_author":"https://pith.science/pith/4H7LDMYF2HZSJ35LMFQ2KYRJDN/action/author_attestation","sign_citation":"https://pith.science/pith/4H7LDMYF2HZSJ35LMFQ2KYRJDN/action/citation_signature","submit_replication":"https://pith.science/pith/4H7LDMYF2HZSJ35LMFQ2KYRJDN/action/replication_record"}},"created_at":"2026-07-05T11:10:42.750464+00:00","updated_at":"2026-07-05T11:10:42.750464+00:00"}