{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:55D3N7NH75VAPZTJKMH3CII4JX","short_pith_number":"pith:55D3N7NH","schema_version":"1.0","canonical_sha256":"ef47b6fda7ff6a07e669530fb1211c4dea7f8dee9f172b3a5ec6a6605120c980","source":{"kind":"arxiv","id":"2210.01241","version":3},"attestation_state":"computed","paper":{"title":"Is Reinforcement Learning (Not) for Natural Language Processing: Benchmarks, Baselines, and Building Blocks for Natural Language Policy Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Christian Bauckhage, Hannaneh Hajishirzi, Jack Hessel, Kiant\\'e Brantley, Prithviraj Ammanabrolu, Rafet Sifa, Rajkumar Ramamurthy, Yejin Choi","submitted_at":"2022-10-03T21:38:29Z","abstract_excerpt":"We tackle the problem of aligning pre-trained large language models (LMs) with human preferences. If we view text generation as a sequential decision-making problem, reinforcement learning (RL) appears to be a natural conceptual framework. However, using RL for LM-based generation faces empirical challenges, including training instability due to the combinatorial action space, as well as a lack of open-source libraries and benchmarks customized for LM alignment. Thus, a question rises in the research community: is RL a practical paradigm for NLP?\n  To help answer this, we first introduce an op"},"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":"2210.01241","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-10-03T21:38:29Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"74bd7df0b4a90877838254155174abb485c17553e1b4f9552882117ecbff9dda","abstract_canon_sha256":"3d8885c3ea7a90d7f929dacf2aa09d05651af47f0050831c3b7604ec8b10bc3b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:19:41.845338Z","signature_b64":"3Q/S6tx+cRgD/pMSsGuqQmJbWXGJVH1//lr/wwKfGrOsoc4wRFp0jkH6XMcfgfe5yZE4AVTBIBQOxPri8QD3BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ef47b6fda7ff6a07e669530fb1211c4dea7f8dee9f172b3a5ec6a6605120c980","last_reissued_at":"2026-07-05T07:19:41.844857Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:19:41.844857Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Is Reinforcement Learning (Not) for Natural Language Processing: Benchmarks, Baselines, and Building Blocks for Natural Language Policy Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Christian Bauckhage, Hannaneh Hajishirzi, Jack Hessel, Kiant\\'e Brantley, Prithviraj Ammanabrolu, Rafet Sifa, Rajkumar Ramamurthy, Yejin Choi","submitted_at":"2022-10-03T21:38:29Z","abstract_excerpt":"We tackle the problem of aligning pre-trained large language models (LMs) with human preferences. If we view text generation as a sequential decision-making problem, reinforcement learning (RL) appears to be a natural conceptual framework. However, using RL for LM-based generation faces empirical challenges, including training instability due to the combinatorial action space, as well as a lack of open-source libraries and benchmarks customized for LM alignment. Thus, a question rises in the research community: is RL a practical paradigm for NLP?\n  To help answer this, we first introduce an op"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.01241","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/2210.01241/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":"2210.01241","created_at":"2026-07-05T07:19:41.844915+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.01241v3","created_at":"2026-07-05T07:19:41.844915+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.01241","created_at":"2026-07-05T07:19:41.844915+00:00"},{"alias_kind":"pith_short_12","alias_value":"55D3N7NH75VA","created_at":"2026-07-05T07:19:41.844915+00:00"},{"alias_kind":"pith_short_16","alias_value":"55D3N7NH75VAPZTJ","created_at":"2026-07-05T07:19:41.844915+00:00"},{"alias_kind":"pith_short_8","alias_value":"55D3N7NH","created_at":"2026-07-05T07:19:41.844915+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2304.06767","citing_title":"RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment","ref_index":136,"is_internal_anchor":false},{"citing_arxiv_id":"2501.09686","citing_title":"Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Models","ref_index":116,"is_internal_anchor":false},{"citing_arxiv_id":"2603.12510","citing_title":"Red-Teaming Vision-Language-Action Models via Quality Diversity Prompt Generation for Robust Robot Policies","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08926","citing_title":"Bridging SFT and RL: Dynamic Policy Optimization for Robust Reasoning","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06987","citing_title":"Response Time Enhances Alignment with Heterogeneous Preferences","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17475","citing_title":"Waking Up Blind: Cold-Start Optimization of Supervision-Free Agentic Trajectories for Grounded Visual Perception","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/55D3N7NH75VAPZTJKMH3CII4JX","json":"https://pith.science/pith/55D3N7NH75VAPZTJKMH3CII4JX.json","graph_json":"https://pith.science/api/pith-number/55D3N7NH75VAPZTJKMH3CII4JX/graph.json","events_json":"https://pith.science/api/pith-number/55D3N7NH75VAPZTJKMH3CII4JX/events.json","paper":"https://pith.science/paper/55D3N7NH"},"agent_actions":{"view_html":"https://pith.science/pith/55D3N7NH75VAPZTJKMH3CII4JX","download_json":"https://pith.science/pith/55D3N7NH75VAPZTJKMH3CII4JX.json","view_paper":"https://pith.science/paper/55D3N7NH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.01241&json=true","fetch_graph":"https://pith.science/api/pith-number/55D3N7NH75VAPZTJKMH3CII4JX/graph.json","fetch_events":"https://pith.science/api/pith-number/55D3N7NH75VAPZTJKMH3CII4JX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/55D3N7NH75VAPZTJKMH3CII4JX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/55D3N7NH75VAPZTJKMH3CII4JX/action/storage_attestation","attest_author":"https://pith.science/pith/55D3N7NH75VAPZTJKMH3CII4JX/action/author_attestation","sign_citation":"https://pith.science/pith/55D3N7NH75VAPZTJKMH3CII4JX/action/citation_signature","submit_replication":"https://pith.science/pith/55D3N7NH75VAPZTJKMH3CII4JX/action/replication_record"}},"created_at":"2026-07-05T07:19:41.844915+00:00","updated_at":"2026-07-05T07:19:41.844915+00:00"}