{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ZLKNRCPK4MAAJTZ7IFB5LDAWON","short_pith_number":"pith:ZLKNRCPK","schema_version":"1.0","canonical_sha256":"cad4d889eae30004cf3f4143d58c167378a27d44df7edae3081ee47bd6b4c862","source":{"kind":"arxiv","id":"2411.05273","version":2},"attestation_state":"computed","paper":{"title":"Real-World Offline Reinforcement Learning from Vision Language Model Feedback","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"David Held, Navin Sriram Ravie, Sreyas Venkataraman, Yufei Wang, Zackory Erickson, Ziyu Wang","submitted_at":"2024-11-08T02:12:34Z","abstract_excerpt":"Offline reinforcement learning can enable policy learning from pre-collected, sub-optimal datasets without online interactions. This makes it ideal for real-world robots and safety-critical scenarios, where collecting online data or expert demonstrations is slow, costly, and risky. However, most existing offline RL works assume the dataset is already labeled with the task rewards, a process that often requires significant human effort, especially when ground-truth states are hard to ascertain (e.g., in the real-world). In this paper, we build on prior work, specifically RL-VLM-F, and propose a"},"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":"2411.05273","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2024-11-08T02:12:34Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"0963e08d11ae67471061ede16b1314bcee2e79ea78263a6a6b5adf61f51fc087","abstract_canon_sha256":"90203bb472ccf95dfc93cf221b7643d2228cec1eb7baa587844fc1906dc9e501"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:49:05.230027Z","signature_b64":"DNTb5s6QXKvHSjP4QkH/PCms9vU7b2NoxaUFoeN/Y72UaQVhFTR85YeBJ+3hQgioahknTlBtjPn4JC0nmtqFBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cad4d889eae30004cf3f4143d58c167378a27d44df7edae3081ee47bd6b4c862","last_reissued_at":"2026-07-05T11:49:05.229520Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:49:05.229520Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Real-World Offline Reinforcement Learning from Vision Language Model Feedback","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"David Held, Navin Sriram Ravie, Sreyas Venkataraman, Yufei Wang, Zackory Erickson, Ziyu Wang","submitted_at":"2024-11-08T02:12:34Z","abstract_excerpt":"Offline reinforcement learning can enable policy learning from pre-collected, sub-optimal datasets without online interactions. This makes it ideal for real-world robots and safety-critical scenarios, where collecting online data or expert demonstrations is slow, costly, and risky. However, most existing offline RL works assume the dataset is already labeled with the task rewards, a process that often requires significant human effort, especially when ground-truth states are hard to ascertain (e.g., in the real-world). In this paper, we build on prior work, specifically RL-VLM-F, and propose a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.05273","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/2411.05273/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":"2411.05273","created_at":"2026-07-05T11:49:05.229584+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.05273v2","created_at":"2026-07-05T11:49:05.229584+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.05273","created_at":"2026-07-05T11:49:05.229584+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZLKNRCPK4MAA","created_at":"2026-07-05T11:49:05.229584+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZLKNRCPK4MAAJTZ7","created_at":"2026-07-05T11:49:05.229584+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZLKNRCPK","created_at":"2026-07-05T11:49:05.229584+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23640","citing_title":"Learning Process Rewards via Success Visitation Matching for Efficient RL","ref_index":84,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22123","citing_title":"Beyond Pixels: Learning Invariant Rewards for Real-World Robotics From a Few Demonstrations","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZLKNRCPK4MAAJTZ7IFB5LDAWON","json":"https://pith.science/pith/ZLKNRCPK4MAAJTZ7IFB5LDAWON.json","graph_json":"https://pith.science/api/pith-number/ZLKNRCPK4MAAJTZ7IFB5LDAWON/graph.json","events_json":"https://pith.science/api/pith-number/ZLKNRCPK4MAAJTZ7IFB5LDAWON/events.json","paper":"https://pith.science/paper/ZLKNRCPK"},"agent_actions":{"view_html":"https://pith.science/pith/ZLKNRCPK4MAAJTZ7IFB5LDAWON","download_json":"https://pith.science/pith/ZLKNRCPK4MAAJTZ7IFB5LDAWON.json","view_paper":"https://pith.science/paper/ZLKNRCPK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.05273&json=true","fetch_graph":"https://pith.science/api/pith-number/ZLKNRCPK4MAAJTZ7IFB5LDAWON/graph.json","fetch_events":"https://pith.science/api/pith-number/ZLKNRCPK4MAAJTZ7IFB5LDAWON/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZLKNRCPK4MAAJTZ7IFB5LDAWON/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZLKNRCPK4MAAJTZ7IFB5LDAWON/action/storage_attestation","attest_author":"https://pith.science/pith/ZLKNRCPK4MAAJTZ7IFB5LDAWON/action/author_attestation","sign_citation":"https://pith.science/pith/ZLKNRCPK4MAAJTZ7IFB5LDAWON/action/citation_signature","submit_replication":"https://pith.science/pith/ZLKNRCPK4MAAJTZ7IFB5LDAWON/action/replication_record"}},"created_at":"2026-07-05T11:49:05.229584+00:00","updated_at":"2026-07-05T11:49:05.229584+00:00"}