{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QBB22QNJJBN3DXHYCIABH3IZTF","short_pith_number":"pith:QBB22QNJ","schema_version":"1.0","canonical_sha256":"8043ad41a9485bb1dcf8120013ed199954d162ae158ab4831346f6a2c4208df2","source":{"kind":"arxiv","id":"2411.14251","version":3},"attestation_state":"computed","paper":{"title":"Natural Language Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Bo Liu, Girish A. Koushik, Haotian Fu, Jun Wang, Mengyue Yang, Xidong Feng, Yan Song, Ying Wen, Zhiyuan Hu, Ziyu Wan","submitted_at":"2024-11-21T15:57:02Z","abstract_excerpt":"Artificial intelligence progresses towards the \"Era of Experience,\" where agents are expected to learn from continuous, grounded interaction. We argue that traditional Reinforcement Learning (RL), which typically represents value as a scalar, can restrict agent's deep understanding of environments and hinders the active, deliberative learning crucial for navigating this new paradigm. To address the issue, we introduce Natural Language Reinforcement Learning (NLRL), a framework that extends RL principles into natural language counterparts. Central to NLRL is the Language Value Function (LVF), w"},"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":"2411.14251","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-21T15:57:02Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"2730f12c8ea64b6b5a2cf8d1b0e0e2766b970f0598984d8a414a930985835a6f","abstract_canon_sha256":"50fbf9e03d93d387d1831b6d2b1983b9ccba73f6f511755183457e59f79b9a8e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:10:48.990676Z","signature_b64":"yieXrR6XnDNeY649PvdotLD2VRXbVbNFyNLuFeo6vdUNjAJrPIr3YT2jBvSamMZ8CZZ6UdkxRKJk1NqbCU1aBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8043ad41a9485bb1dcf8120013ed199954d162ae158ab4831346f6a2c4208df2","last_reissued_at":"2026-07-05T11:10:48.990185Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:10:48.990185Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Natural Language Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Bo Liu, Girish A. Koushik, Haotian Fu, Jun Wang, Mengyue Yang, Xidong Feng, Yan Song, Ying Wen, Zhiyuan Hu, Ziyu Wan","submitted_at":"2024-11-21T15:57:02Z","abstract_excerpt":"Artificial intelligence progresses towards the \"Era of Experience,\" where agents are expected to learn from continuous, grounded interaction. We argue that traditional Reinforcement Learning (RL), which typically represents value as a scalar, can restrict agent's deep understanding of environments and hinders the active, deliberative learning crucial for navigating this new paradigm. To address the issue, we introduce Natural Language Reinforcement Learning (NLRL), a framework that extends RL principles into natural language counterparts. Central to NLRL is the Language Value Function (LVF), w"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.14251","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/2411.14251/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":"2411.14251","created_at":"2026-07-05T11:10:48.990244+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.14251v3","created_at":"2026-07-05T11:10:48.990244+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.14251","created_at":"2026-07-05T11:10:48.990244+00:00"},{"alias_kind":"pith_short_12","alias_value":"QBB22QNJJBN3","created_at":"2026-07-05T11:10:48.990244+00:00"},{"alias_kind":"pith_short_16","alias_value":"QBB22QNJJBN3DXHY","created_at":"2026-07-05T11:10:48.990244+00:00"},{"alias_kind":"pith_short_8","alias_value":"QBB22QNJ","created_at":"2026-07-05T11:10:48.990244+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.04389","citing_title":"When Clients Stop Following: A Cognitive Conceptualization Diagram-driven Framework for Strategic Counseling","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15113","citing_title":"Learning from Language Feedback via Variational Policy Distillation","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QBB22QNJJBN3DXHYCIABH3IZTF","json":"https://pith.science/pith/QBB22QNJJBN3DXHYCIABH3IZTF.json","graph_json":"https://pith.science/api/pith-number/QBB22QNJJBN3DXHYCIABH3IZTF/graph.json","events_json":"https://pith.science/api/pith-number/QBB22QNJJBN3DXHYCIABH3IZTF/events.json","paper":"https://pith.science/paper/QBB22QNJ"},"agent_actions":{"view_html":"https://pith.science/pith/QBB22QNJJBN3DXHYCIABH3IZTF","download_json":"https://pith.science/pith/QBB22QNJJBN3DXHYCIABH3IZTF.json","view_paper":"https://pith.science/paper/QBB22QNJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.14251&json=true","fetch_graph":"https://pith.science/api/pith-number/QBB22QNJJBN3DXHYCIABH3IZTF/graph.json","fetch_events":"https://pith.science/api/pith-number/QBB22QNJJBN3DXHYCIABH3IZTF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QBB22QNJJBN3DXHYCIABH3IZTF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QBB22QNJJBN3DXHYCIABH3IZTF/action/storage_attestation","attest_author":"https://pith.science/pith/QBB22QNJJBN3DXHYCIABH3IZTF/action/author_attestation","sign_citation":"https://pith.science/pith/QBB22QNJJBN3DXHYCIABH3IZTF/action/citation_signature","submit_replication":"https://pith.science/pith/QBB22QNJJBN3DXHYCIABH3IZTF/action/replication_record"}},"created_at":"2026-07-05T11:10:48.990244+00:00","updated_at":"2026-07-05T11:10:48.990244+00:00"}