{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:DHNPYQ5G2DWUV3T2BAPT6HOBNY","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"6438bc3f9b0ad57edc0c56bf97ba63f2ed3cd3b4b78e640459a294c532011eea","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-29T09:24:30Z","title_canon_sha256":"a3ee8a00abb8653ecd09efee5e6e3dff17022749f7ceb3f9a19d8760fc0ac677"},"schema_version":"1.0","source":{"id":"2504.20571","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.20571","created_at":"2026-05-17T23:38:50Z"},{"alias_kind":"arxiv_version","alias_value":"2504.20571v3","created_at":"2026-05-17T23:38:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.20571","created_at":"2026-05-17T23:38:50Z"},{"alias_kind":"pith_short_12","alias_value":"DHNPYQ5G2DWU","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"DHNPYQ5G2DWUV3T2","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"DHNPYQ5G","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:2259304bd3fb97791c16045f1a6572b856a2406807eb9cebb0bfa183b2d5b828","target":"graph","created_at":"2026-05-17T23:38:50Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Reinforcement learning with verifiable reward using one training example (1-shot RLVR) is effective in incentivizing the math reasoning capabilities of large language models (LLMs). Applying RLVR to the base model Qwen2.5-Math-1.5B, we identify a single example that elevates model performance on MATH500 from 36.0% to 73.6%."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the single chosen example is not specially selected in a way that inflates generalization, and that the observed gains arise from the RL policy gradient rather than incidental effects of the training setup or prompt format."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"One training example via RLVR boosts LLM math reasoning from 17.6% to 35.7% average across six benchmarks."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"One training example via reinforcement learning lifts an LLM's math reasoning score from 36% to 74% on MATH500."}],"snapshot_sha256":"36735173058404e2fa8e7aad18c2de165a2691a58c91a23b42e5aca735632a56"},"formal_canon":{"evidence_count":1,"snapshot_sha256":"dbbdd7d28668dbf699e7d875b7e1415989d324785463f554cf12a2052ab56764"},"paper":{"abstract_excerpt":"We show that reinforcement learning with verifiable reward using one training example (1-shot RLVR) is effective in incentivizing the math reasoning capabilities of large language models (LLMs). Applying RLVR to the base model Qwen2.5-Math-1.5B, we identify a single example that elevates model performance on MATH500 from 36.0% to 73.6% (8.6% improvement beyond format correction), and improves the average performance across six common mathematical reasoning benchmarks from 17.6% to 35.7% (7.0% non-format gain). This result matches the performance obtained using the 1.2k DeepScaleR subset (MATH5","authors_text":"Baolin Peng, Hao Cheng, Jianfeng Gao, Kuan Wang, Liliang Ren, Liyuan Liu, Qing Yang, Shuohang Wang, Simon Shaolei Du, Weizhu Chen, Xuehai He, Yelong Shen, Yiping Wang, Zhiyuan Zeng","cross_cats":["cs.AI","cs.CL"],"headline":"One training example via reinforcement learning lifts an LLM's math reasoning score from 36% to 74% on MATH500.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-29T09:24:30Z","title":"Reinforcement Learning for Reasoning in Large Language Models with One Training Example"},"references":{"count":69,"internal_anchors":29,"resolved_work":69,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Learning to reason with llms","work_id":"170ab9a7-f02d-4838-a217-516dcecfaaf1","year":2024},{"cited_arxiv_id":"2501.12948","doi":"","is_internal_anchor":true,"ref_index":2,"title":"DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning","work_id":"e6b75ad5-2877-4168-97c8-710407094d20","year":2025},{"cited_arxiv_id":"2501.12599","doi":"","is_internal_anchor":true,"ref_index":3,"title":"Kimi k1.5: Scaling Reinforcement Learning with LLMs","work_id":"bff96ab1-bd6a-4585-be23-74fdb51969c7","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"On designing effective rl reward at training time for llm reasoning","work_id":"5c46cc3b-db37-4655-a0ff-f28ab6c32a1f","year":2024},{"cited_arxiv_id":"2411.15124","doi":"","is_internal_anchor":true,"ref_index":5,"title":"Tulu 3: Pushing Frontiers in Open Language Model Post-Training","work_id":"28c9dbea-056a-48c2-8000-85f809827e45","year":2024}],"snapshot_sha256":"57a5e9920dcc9893c568b300a373c6224244244cb7150ad9389f9bf9eeae1bad"},"source":{"id":"2504.20571","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-15T19:47:48.868083Z","id":"c7b86817-aa42-4484-a016-6876f0248f25","model_set":{"reader":"grok-4.3"},"one_line_summary":"One training example via RLVR boosts LLM math reasoning from 17.6% to 35.7% average across six benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"One training example via reinforcement learning lifts an LLM's math reasoning score from 36% to 74% on MATH500.","strongest_claim":"Reinforcement learning with verifiable reward using one training example (1-shot RLVR) is effective in incentivizing the math reasoning capabilities of large language models (LLMs). Applying RLVR to the base model Qwen2.5-Math-1.5B, we identify a single example that elevates model performance on MATH500 from 36.0% to 73.6%.","weakest_assumption":"That the single chosen example is not specially selected in a way that inflates generalization, and that the observed gains arise from the RL policy gradient rather than incidental effects of the training setup or prompt format."}},"verdict_id":"c7b86817-aa42-4484-a016-6876f0248f25"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:f6ac5fc21967e219b9a37db0619e657ccfc29ede9c2c1d0f6c9b3e9f76f2b94e","target":"record","created_at":"2026-05-17T23:38:50Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"6438bc3f9b0ad57edc0c56bf97ba63f2ed3cd3b4b78e640459a294c532011eea","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-29T09:24:30Z","title_canon_sha256":"a3ee8a00abb8653ecd09efee5e6e3dff17022749f7ceb3f9a19d8760fc0ac677"},"schema_version":"1.0","source":{"id":"2504.20571","kind":"arxiv","version":3}},"canonical_sha256":"19dafc43a6d0ed4aee7a081f3f1dc16e2690bce9e729cb67e4cf0fedc999d5d7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"19dafc43a6d0ed4aee7a081f3f1dc16e2690bce9e729cb67e4cf0fedc999d5d7","first_computed_at":"2026-05-17T23:38:50.357211Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:38:50.357211Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"CDcBAeWaHXOcNc6o8DYNgDNh+UFk5jeLsgAZXk00XQtDJf8k+rul2TPAweNEkphMes3LZWedgsqG0QSWrWZoAw==","signature_status":"signed_v1","signed_at":"2026-05-17T23:38:50.357719Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.20571","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f6ac5fc21967e219b9a37db0619e657ccfc29ede9c2c1d0f6c9b3e9f76f2b94e","sha256:2259304bd3fb97791c16045f1a6572b856a2406807eb9cebb0bfa183b2d5b828"],"state_sha256":"c5054a626d78596ff4ba47d8ae385e8ab7113c9bb75d876471866dd768681b24"}