{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:AOCPRYJ4SF35X7HACGA2NDUFQQ","short_pith_number":"pith:AOCPRYJ4","schema_version":"1.0","canonical_sha256":"0384f8e13c9177dbfce01181a68e85842491f073fce8ee6b3ae7b74569728438","source":{"kind":"arxiv","id":"2506.08007","version":1},"attestation_state":"computed","paper":{"title":"Reinforcement Pre-Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Furu Wei, Li Dong, Qingxiu Dong, Tianzhu Ye, Yao Tang, Yutao Sun, Zhifang Sui","submitted_at":"2025-06-09T17:59:53Z","abstract_excerpt":"In this work, we introduce Reinforcement Pre-Training (RPT) as a new scaling paradigm for large language models and reinforcement learning (RL). Specifically, we reframe next-token prediction as a reasoning task trained using RL, where it receives verifiable rewards for correctly predicting the next token for a given context. RPT offers a scalable method to leverage vast amounts of text data for general-purpose RL, rather than relying on domain-specific annotated answers. By incentivizing the capability of next-token reasoning, RPT significantly improves the language modeling accuracy of predi"},"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":"2506.08007","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-09T17:59:53Z","cross_cats_sorted":[],"title_canon_sha256":"cdff2fe0d2cf952940426e674770674b305393f496c68cb09578d178eb48e3e2","abstract_canon_sha256":"1a9f140ac4cdce90ad7aea99979e5f4efda52812208cbbe4a8da9684450e9172"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:18:37.155177Z","signature_b64":"qGHr4X3S9yPEmQCm3ICPZeKGBY8rpabvMGYDsMNBrXrnaiYIkzU4Yh0Heeulw9Y+qwwBZ8FtX2sraDYQdGe2DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0384f8e13c9177dbfce01181a68e85842491f073fce8ee6b3ae7b74569728438","last_reissued_at":"2026-07-05T11:18:37.154688Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:18:37.154688Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reinforcement Pre-Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Furu Wei, Li Dong, Qingxiu Dong, Tianzhu Ye, Yao Tang, Yutao Sun, Zhifang Sui","submitted_at":"2025-06-09T17:59:53Z","abstract_excerpt":"In this work, we introduce Reinforcement Pre-Training (RPT) as a new scaling paradigm for large language models and reinforcement learning (RL). Specifically, we reframe next-token prediction as a reasoning task trained using RL, where it receives verifiable rewards for correctly predicting the next token for a given context. RPT offers a scalable method to leverage vast amounts of text data for general-purpose RL, rather than relying on domain-specific annotated answers. By incentivizing the capability of next-token reasoning, RPT significantly improves the language modeling accuracy of predi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.08007","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/2506.08007/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":"2506.08007","created_at":"2026-07-05T11:18:37.154746+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.08007v1","created_at":"2026-07-05T11:18:37.154746+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.08007","created_at":"2026-07-05T11:18:37.154746+00:00"},{"alias_kind":"pith_short_12","alias_value":"AOCPRYJ4SF35","created_at":"2026-07-05T11:18:37.154746+00:00"},{"alias_kind":"pith_short_16","alias_value":"AOCPRYJ4SF35X7HA","created_at":"2026-07-05T11:18:37.154746+00:00"},{"alias_kind":"pith_short_8","alias_value":"AOCPRYJ4","created_at":"2026-07-05T11:18:37.154746+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":11,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18961","citing_title":"Be Your Own Teacher: Steering Protein Language Models via Unsupervised Reward Optimization","ref_index":59,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04272","citing_title":"RL Excursions during Pre-Training: Re-examining Policy Optimization for LLM training","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01249","citing_title":"Trust Region On-Policy Distillation","ref_index":82,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21654","citing_title":"Value-Gradient Hypothesis of RL for LLMs","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19444","citing_title":"Detecting and Mitigating the Correct-Answer Extinction Window in Test-Time Reinforcement Learning with Majority Voting","ref_index":29,"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":14,"is_internal_anchor":false},{"citing_arxiv_id":"2509.02547","citing_title":"The Landscape of Agentic Reinforcement Learning for LLMs: A Survey","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2509.08827","citing_title":"A Survey of Reinforcement Learning for Large Reasoning Models","ref_index":116,"is_internal_anchor":false},{"citing_arxiv_id":"2601.12538","citing_title":"Agentic Reasoning for Large Language Models","ref_index":207,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14142","citing_title":"From $P(y|x)$ to $P(y)$: Investigating Reinforcement Learning in Pre-train Space","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12160","citing_title":"PubSwap: Public-Data Off-Policy Coordination for Federated RLVR","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AOCPRYJ4SF35X7HACGA2NDUFQQ","json":"https://pith.science/pith/AOCPRYJ4SF35X7HACGA2NDUFQQ.json","graph_json":"https://pith.science/api/pith-number/AOCPRYJ4SF35X7HACGA2NDUFQQ/graph.json","events_json":"https://pith.science/api/pith-number/AOCPRYJ4SF35X7HACGA2NDUFQQ/events.json","paper":"https://pith.science/paper/AOCPRYJ4"},"agent_actions":{"view_html":"https://pith.science/pith/AOCPRYJ4SF35X7HACGA2NDUFQQ","download_json":"https://pith.science/pith/AOCPRYJ4SF35X7HACGA2NDUFQQ.json","view_paper":"https://pith.science/paper/AOCPRYJ4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.08007&json=true","fetch_graph":"https://pith.science/api/pith-number/AOCPRYJ4SF35X7HACGA2NDUFQQ/graph.json","fetch_events":"https://pith.science/api/pith-number/AOCPRYJ4SF35X7HACGA2NDUFQQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AOCPRYJ4SF35X7HACGA2NDUFQQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AOCPRYJ4SF35X7HACGA2NDUFQQ/action/storage_attestation","attest_author":"https://pith.science/pith/AOCPRYJ4SF35X7HACGA2NDUFQQ/action/author_attestation","sign_citation":"https://pith.science/pith/AOCPRYJ4SF35X7HACGA2NDUFQQ/action/citation_signature","submit_replication":"https://pith.science/pith/AOCPRYJ4SF35X7HACGA2NDUFQQ/action/replication_record"}},"created_at":"2026-07-05T11:18:37.154746+00:00","updated_at":"2026-07-05T11:18:37.154746+00:00"}