{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:NNDJBHYX2NAPF4JNXLXA2UNKM2","short_pith_number":"pith:NNDJBHYX","schema_version":"1.0","canonical_sha256":"6b46909f17d340f2f12dbaee0d51aa66a8b21426e25cd9adb84a04465edec5b6","source":{"kind":"arxiv","id":"2402.07876","version":6},"attestation_state":"computed","paper":{"title":"Policy Improvement using Language Feedback Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Dipendra Misra, Marc-Alexandre C\\^ot\\'e, Victor Zhong, Xingdi Yuan","submitted_at":"2024-02-12T18:41:34Z","abstract_excerpt":"We introduce Language Feedback Models (LFMs) that identify desirable behaviour - actions that help achieve tasks specified in the instruction - for imitation learning in instruction following. To train LFMs, we obtain feedback from Large Language Models (LLMs) on visual trajectories verbalized to language descriptions. First, by using LFMs to identify desirable behaviour to imitate, we improve in task-completion rate over strong behavioural cloning baselines on three distinct language grounding environments (Touchdown, ScienceWorld, and ALFWorld). Second, LFMs outperform using LLMs as experts "},"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":"2402.07876","kind":"arxiv","version":6},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-12T18:41:34Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"6fad1ad812949a1b6b11081376e986472efb510eb27b1f879365d35c083d146b","abstract_canon_sha256":"776f0d0e8712177ae9b04011e39dc56fe9b385be1038bc99871916f739e4550d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:18:22.918370Z","signature_b64":"tW0PXY5avWYhbWk9yVVT6XDjY7ofajMF+xc8jyonriTJOnbyxOuxxGDOvKBpPaYGB49XQB1hY38NULfpdD6cAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6b46909f17d340f2f12dbaee0d51aa66a8b21426e25cd9adb84a04465edec5b6","last_reissued_at":"2026-07-05T09:18:22.917863Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:18:22.917863Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Policy Improvement using Language Feedback Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Dipendra Misra, Marc-Alexandre C\\^ot\\'e, Victor Zhong, Xingdi Yuan","submitted_at":"2024-02-12T18:41:34Z","abstract_excerpt":"We introduce Language Feedback Models (LFMs) that identify desirable behaviour - actions that help achieve tasks specified in the instruction - for imitation learning in instruction following. To train LFMs, we obtain feedback from Large Language Models (LLMs) on visual trajectories verbalized to language descriptions. First, by using LFMs to identify desirable behaviour to imitate, we improve in task-completion rate over strong behavioural cloning baselines on three distinct language grounding environments (Touchdown, ScienceWorld, and ALFWorld). Second, LFMs outperform using LLMs as experts "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.07876","kind":"arxiv","version":6},"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/2402.07876/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":"2402.07876","created_at":"2026-07-05T09:18:22.917920+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.07876v6","created_at":"2026-07-05T09:18:22.917920+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.07876","created_at":"2026-07-05T09:18:22.917920+00:00"},{"alias_kind":"pith_short_12","alias_value":"NNDJBHYX2NAP","created_at":"2026-07-05T09:18:22.917920+00:00"},{"alias_kind":"pith_short_16","alias_value":"NNDJBHYX2NAPF4JN","created_at":"2026-07-05T09:18:22.917920+00:00"},{"alias_kind":"pith_short_8","alias_value":"NNDJBHYX","created_at":"2026-07-05T09:18:22.917920+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.10341","citing_title":"Formalizing Learning from Language Feedback with Provable Guarantees","ref_index":39,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NNDJBHYX2NAPF4JNXLXA2UNKM2","json":"https://pith.science/pith/NNDJBHYX2NAPF4JNXLXA2UNKM2.json","graph_json":"https://pith.science/api/pith-number/NNDJBHYX2NAPF4JNXLXA2UNKM2/graph.json","events_json":"https://pith.science/api/pith-number/NNDJBHYX2NAPF4JNXLXA2UNKM2/events.json","paper":"https://pith.science/paper/NNDJBHYX"},"agent_actions":{"view_html":"https://pith.science/pith/NNDJBHYX2NAPF4JNXLXA2UNKM2","download_json":"https://pith.science/pith/NNDJBHYX2NAPF4JNXLXA2UNKM2.json","view_paper":"https://pith.science/paper/NNDJBHYX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.07876&json=true","fetch_graph":"https://pith.science/api/pith-number/NNDJBHYX2NAPF4JNXLXA2UNKM2/graph.json","fetch_events":"https://pith.science/api/pith-number/NNDJBHYX2NAPF4JNXLXA2UNKM2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NNDJBHYX2NAPF4JNXLXA2UNKM2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NNDJBHYX2NAPF4JNXLXA2UNKM2/action/storage_attestation","attest_author":"https://pith.science/pith/NNDJBHYX2NAPF4JNXLXA2UNKM2/action/author_attestation","sign_citation":"https://pith.science/pith/NNDJBHYX2NAPF4JNXLXA2UNKM2/action/citation_signature","submit_replication":"https://pith.science/pith/NNDJBHYX2NAPF4JNXLXA2UNKM2/action/replication_record"}},"created_at":"2026-07-05T09:18:22.917920+00:00","updated_at":"2026-07-05T09:18:22.917920+00:00"}