{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:QDHBP43TIUURC2QY6M5P52GP7R","short_pith_number":"pith:QDHBP43T","schema_version":"1.0","canonical_sha256":"80ce17f3734529116a18f33afee8cffc4713f4d54bcae9d50dbcf48154fb999b","source":{"kind":"arxiv","id":"2507.23391","version":1},"attestation_state":"computed","paper":{"title":"Policy Learning from Large Vision-Language Model Feedback without Reward Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.LG","authors_text":"Chang D. Yoo, Donghoon Lee, Tung M. Luu, Younghwan Lee","submitted_at":"2025-07-31T10:07:49Z","abstract_excerpt":"Offline reinforcement learning (RL) provides a powerful framework for training robotic agents using pre-collected, suboptimal datasets, eliminating the need for costly, time-consuming, and potentially hazardous online interactions. This is particularly useful in safety-critical real-world applications, where online data collection is expensive and impractical. However, existing offline RL algorithms typically require reward labeled data, which introduces an additional bottleneck: reward function design is itself costly, labor-intensive, and requires significant domain expertise. In this paper,"},"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":"2507.23391","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-31T10:07:49Z","cross_cats_sorted":["cs.RO"],"title_canon_sha256":"a3e53122af9f199a7345ee6fda2225ee332164307d90189105900a33f007d6a9","abstract_canon_sha256":"0ecef0af0fae645a2f41f76a3250f8e90d8c9e37983dea139e899ae81df7ef9b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:46:11.321196Z","signature_b64":"jxSJ4Q/0oQB9daTpR/jOORQm6gFGf2ICHkc0935VIZspe2Kz0Z0l31FBbcTVPV9AxC6wKMvUbjgEOpEQZ3axAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"80ce17f3734529116a18f33afee8cffc4713f4d54bcae9d50dbcf48154fb999b","last_reissued_at":"2026-07-05T11:46:11.320777Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:46:11.320777Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Policy Learning from Large Vision-Language Model Feedback without Reward Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.LG","authors_text":"Chang D. Yoo, Donghoon Lee, Tung M. Luu, Younghwan Lee","submitted_at":"2025-07-31T10:07:49Z","abstract_excerpt":"Offline reinforcement learning (RL) provides a powerful framework for training robotic agents using pre-collected, suboptimal datasets, eliminating the need for costly, time-consuming, and potentially hazardous online interactions. This is particularly useful in safety-critical real-world applications, where online data collection is expensive and impractical. However, existing offline RL algorithms typically require reward labeled data, which introduces an additional bottleneck: reward function design is itself costly, labor-intensive, and requires significant domain expertise. In this paper,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.23391","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/2507.23391/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":"2507.23391","created_at":"2026-07-05T11:46:11.320839+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.23391v1","created_at":"2026-07-05T11:46:11.320839+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.23391","created_at":"2026-07-05T11:46:11.320839+00:00"},{"alias_kind":"pith_short_12","alias_value":"QDHBP43TIUUR","created_at":"2026-07-05T11:46:11.320839+00:00"},{"alias_kind":"pith_short_16","alias_value":"QDHBP43TIUURC2QY","created_at":"2026-07-05T11:46:11.320839+00:00"},{"alias_kind":"pith_short_8","alias_value":"QDHBP43T","created_at":"2026-07-05T11:46:11.320839+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QDHBP43TIUURC2QY6M5P52GP7R","json":"https://pith.science/pith/QDHBP43TIUURC2QY6M5P52GP7R.json","graph_json":"https://pith.science/api/pith-number/QDHBP43TIUURC2QY6M5P52GP7R/graph.json","events_json":"https://pith.science/api/pith-number/QDHBP43TIUURC2QY6M5P52GP7R/events.json","paper":"https://pith.science/paper/QDHBP43T"},"agent_actions":{"view_html":"https://pith.science/pith/QDHBP43TIUURC2QY6M5P52GP7R","download_json":"https://pith.science/pith/QDHBP43TIUURC2QY6M5P52GP7R.json","view_paper":"https://pith.science/paper/QDHBP43T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.23391&json=true","fetch_graph":"https://pith.science/api/pith-number/QDHBP43TIUURC2QY6M5P52GP7R/graph.json","fetch_events":"https://pith.science/api/pith-number/QDHBP43TIUURC2QY6M5P52GP7R/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QDHBP43TIUURC2QY6M5P52GP7R/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QDHBP43TIUURC2QY6M5P52GP7R/action/storage_attestation","attest_author":"https://pith.science/pith/QDHBP43TIUURC2QY6M5P52GP7R/action/author_attestation","sign_citation":"https://pith.science/pith/QDHBP43TIUURC2QY6M5P52GP7R/action/citation_signature","submit_replication":"https://pith.science/pith/QDHBP43TIUURC2QY6M5P52GP7R/action/replication_record"}},"created_at":"2026-07-05T11:46:11.320839+00:00","updated_at":"2026-07-05T11:46:11.320839+00:00"}