{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:LXZZ2N4MSOI747P3PWUACNF4MD","short_pith_number":"pith:LXZZ2N4M","schema_version":"1.0","canonical_sha256":"5df39d378c9391fe7dfb7da80134bc60f787b5e5adbefbfdf6d2c849cca864e2","source":{"kind":"arxiv","id":"2405.15821","version":1},"attestation_state":"computed","paper":{"title":"Reinforcing Language Agents via Policy Optimization with Action Decomposition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Jun Wang, Muning Wen, Weinan Zhang, Ying Wen, Ziyu Wan","submitted_at":"2024-05-23T14:01:44Z","abstract_excerpt":"Language models as intelligent agents push the boundaries of sequential decision-making agents but struggle with limited knowledge of environmental dynamics and exponentially huge action space. Recent efforts like GLAM and TWOSOME manually constrain the action space to a restricted subset and employ reinforcement learning to align agents' knowledge with specific environments. However, they overlook fine-grained credit assignments for intra-action tokens, which is essential for efficient language agent optimization, and rely on human's prior knowledge to restrict action space. This paper propos"},"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":"2405.15821","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-05-23T14:01:44Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"7b376f3723d35b0d21e9860057e0c388bb4a24a75a34f736b9e22f34e4ab6ea3","abstract_canon_sha256":"e1d873611dd02e965635563f3bd73470f1771bdb475bf598494de344a5a67812"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:22:57.274678Z","signature_b64":"J/21gIwcwPJz22T4xqZYyDbL2l4TBdC4mzCeCHopBZJArezlhMld98xP/rUeeUGs0y3h5rDHQPQtcx8osISoDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5df39d378c9391fe7dfb7da80134bc60f787b5e5adbefbfdf6d2c849cca864e2","last_reissued_at":"2026-07-05T08:22:57.274085Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:22:57.274085Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reinforcing Language Agents via Policy Optimization with Action Decomposition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Jun Wang, Muning Wen, Weinan Zhang, Ying Wen, Ziyu Wan","submitted_at":"2024-05-23T14:01:44Z","abstract_excerpt":"Language models as intelligent agents push the boundaries of sequential decision-making agents but struggle with limited knowledge of environmental dynamics and exponentially huge action space. Recent efforts like GLAM and TWOSOME manually constrain the action space to a restricted subset and employ reinforcement learning to align agents' knowledge with specific environments. However, they overlook fine-grained credit assignments for intra-action tokens, which is essential for efficient language agent optimization, and rely on human's prior knowledge to restrict action space. This paper propos"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.15821","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/2405.15821/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":"2405.15821","created_at":"2026-07-05T08:22:57.274166+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.15821v1","created_at":"2026-07-05T08:22:57.274166+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.15821","created_at":"2026-07-05T08:22:57.274166+00:00"},{"alias_kind":"pith_short_12","alias_value":"LXZZ2N4MSOI7","created_at":"2026-07-05T08:22:57.274166+00:00"},{"alias_kind":"pith_short_16","alias_value":"LXZZ2N4MSOI747P3","created_at":"2026-07-05T08:22:57.274166+00:00"},{"alias_kind":"pith_short_8","alias_value":"LXZZ2N4M","created_at":"2026-07-05T08:22:57.274166+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.14558","citing_title":"Resolving Action Bottleneck: Agentic Reinforcement Learning Informed by Token-Level Energy","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09459","citing_title":"From Reasoning to Agentic: Credit Assignment in Reinforcement Learning for Large Language Models","ref_index":30,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LXZZ2N4MSOI747P3PWUACNF4MD","json":"https://pith.science/pith/LXZZ2N4MSOI747P3PWUACNF4MD.json","graph_json":"https://pith.science/api/pith-number/LXZZ2N4MSOI747P3PWUACNF4MD/graph.json","events_json":"https://pith.science/api/pith-number/LXZZ2N4MSOI747P3PWUACNF4MD/events.json","paper":"https://pith.science/paper/LXZZ2N4M"},"agent_actions":{"view_html":"https://pith.science/pith/LXZZ2N4MSOI747P3PWUACNF4MD","download_json":"https://pith.science/pith/LXZZ2N4MSOI747P3PWUACNF4MD.json","view_paper":"https://pith.science/paper/LXZZ2N4M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.15821&json=true","fetch_graph":"https://pith.science/api/pith-number/LXZZ2N4MSOI747P3PWUACNF4MD/graph.json","fetch_events":"https://pith.science/api/pith-number/LXZZ2N4MSOI747P3PWUACNF4MD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LXZZ2N4MSOI747P3PWUACNF4MD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LXZZ2N4MSOI747P3PWUACNF4MD/action/storage_attestation","attest_author":"https://pith.science/pith/LXZZ2N4MSOI747P3PWUACNF4MD/action/author_attestation","sign_citation":"https://pith.science/pith/LXZZ2N4MSOI747P3PWUACNF4MD/action/citation_signature","submit_replication":"https://pith.science/pith/LXZZ2N4MSOI747P3PWUACNF4MD/action/replication_record"}},"created_at":"2026-07-05T08:22:57.274166+00:00","updated_at":"2026-07-05T08:22:57.274166+00:00"}