{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QOC75PZFJJ2SJI5ROS4ELKVZV3","short_pith_number":"pith:QOC75PZF","schema_version":"1.0","canonical_sha256":"8385febf254a7524a3b174b845aab9aec3bc38a2a70966f981bf202ba7fe55e2","source":{"kind":"arxiv","id":"2401.14151","version":2},"attestation_state":"computed","paper":{"title":"True Knowledge Comes from Practice: Aligning LLMs with Embodied Environments via Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Bo An, Longtao Zheng, Shanqi Liu, Weihao Tan, Wentao Zhang, Xinrun Wang","submitted_at":"2024-01-25T13:03:20Z","abstract_excerpt":"Despite the impressive performance across numerous tasks, large language models (LLMs) often fail in solving simple decision-making tasks due to the misalignment of the knowledge in LLMs with environments. On the contrary, reinforcement learning (RL) agents learn policies from scratch, which makes them always align with environments but difficult to incorporate prior knowledge for efficient explorations. To narrow the gap, we propose TWOSOME, a novel general online framework that deploys LLMs as decision-making agents to efficiently interact and align with embodied environments via RL without "},"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":"2401.14151","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-01-25T13:03:20Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"712cdf6d3b2998e7d2198c03f7c4dbb4705bb7ecae08dd51aeeef7322e0b1558","abstract_canon_sha256":"d9251d3df6eb0efbba9bf20d79a2cc11fc21b48f4e62f8ee9d835c734173c307"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:54:35.592564Z","signature_b64":"NYd9dzXHJh46CgEYwc7pAY9PGdIi80MokrhYiDlGJmfxRj74uygrMo/whTsnN7kABYB/BJHQO7UrNKk30TK5Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8385febf254a7524a3b174b845aab9aec3bc38a2a70966f981bf202ba7fe55e2","last_reissued_at":"2026-07-05T07:54:35.592087Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:54:35.592087Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"True Knowledge Comes from Practice: Aligning LLMs with Embodied Environments via Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Bo An, Longtao Zheng, Shanqi Liu, Weihao Tan, Wentao Zhang, Xinrun Wang","submitted_at":"2024-01-25T13:03:20Z","abstract_excerpt":"Despite the impressive performance across numerous tasks, large language models (LLMs) often fail in solving simple decision-making tasks due to the misalignment of the knowledge in LLMs with environments. On the contrary, reinforcement learning (RL) agents learn policies from scratch, which makes them always align with environments but difficult to incorporate prior knowledge for efficient explorations. To narrow the gap, we propose TWOSOME, a novel general online framework that deploys LLMs as decision-making agents to efficiently interact and align with embodied environments via RL without "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.14151","kind":"arxiv","version":2},"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/2401.14151/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":"2401.14151","created_at":"2026-07-05T07:54:35.592143+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.14151v2","created_at":"2026-07-05T07:54:35.592143+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.14151","created_at":"2026-07-05T07:54:35.592143+00:00"},{"alias_kind":"pith_short_12","alias_value":"QOC75PZFJJ2S","created_at":"2026-07-05T07:54:35.592143+00:00"},{"alias_kind":"pith_short_16","alias_value":"QOC75PZFJJ2SJI5R","created_at":"2026-07-05T07:54:35.592143+00:00"},{"alias_kind":"pith_short_8","alias_value":"QOC75PZF","created_at":"2026-07-05T07:54:35.592143+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.23330","citing_title":"Structured In-context Environment Scaling for Large Language Model Reasoning","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2603.00977","citing_title":"HiMAC: Hierarchical Macro-Micro Learning for Long-Horizon LLM Agents","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05407","citing_title":"PRISM: Perception Reasoning Interleaved for Sequential Decision Making","ref_index":42,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QOC75PZFJJ2SJI5ROS4ELKVZV3","json":"https://pith.science/pith/QOC75PZFJJ2SJI5ROS4ELKVZV3.json","graph_json":"https://pith.science/api/pith-number/QOC75PZFJJ2SJI5ROS4ELKVZV3/graph.json","events_json":"https://pith.science/api/pith-number/QOC75PZFJJ2SJI5ROS4ELKVZV3/events.json","paper":"https://pith.science/paper/QOC75PZF"},"agent_actions":{"view_html":"https://pith.science/pith/QOC75PZFJJ2SJI5ROS4ELKVZV3","download_json":"https://pith.science/pith/QOC75PZFJJ2SJI5ROS4ELKVZV3.json","view_paper":"https://pith.science/paper/QOC75PZF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.14151&json=true","fetch_graph":"https://pith.science/api/pith-number/QOC75PZFJJ2SJI5ROS4ELKVZV3/graph.json","fetch_events":"https://pith.science/api/pith-number/QOC75PZFJJ2SJI5ROS4ELKVZV3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QOC75PZFJJ2SJI5ROS4ELKVZV3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QOC75PZFJJ2SJI5ROS4ELKVZV3/action/storage_attestation","attest_author":"https://pith.science/pith/QOC75PZFJJ2SJI5ROS4ELKVZV3/action/author_attestation","sign_citation":"https://pith.science/pith/QOC75PZFJJ2SJI5ROS4ELKVZV3/action/citation_signature","submit_replication":"https://pith.science/pith/QOC75PZFJJ2SJI5ROS4ELKVZV3/action/replication_record"}},"created_at":"2026-07-05T07:54:35.592143+00:00","updated_at":"2026-07-05T07:54:35.592143+00:00"}