{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:P6YOQZ2ZGIOBVDSNPZK5HTX45U","short_pith_number":"pith:P6YOQZ2Z","schema_version":"1.0","canonical_sha256":"7fb0e86759321c1a8e4d7e55d3cefced1c78c9124ae4c7b46f21915f2aa01ed4","source":{"kind":"arxiv","id":"2310.00771","version":4},"attestation_state":"computed","paper":{"title":"Pre-training with Synthetic Data Helps Offline Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Che Wang, Keith Ross, Zecheng Wang, Zixuan Dong","submitted_at":"2023-10-01T19:32:14Z","abstract_excerpt":"Recently, it has been shown that for offline deep reinforcement learning (DRL), pre-training Decision Transformer with a large language corpus can improve downstream performance (Reid et al., 2022). A natural question to ask is whether this performance gain can only be achieved with language pre-training, or can be achieved with simpler pre-training schemes which do not involve language. In this paper, we first show that language is not essential for improved performance, and indeed pre-training with synthetic IID data for a small number of updates can match the performance gains from pre-trai"},"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":"2310.00771","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2023-10-01T19:32:14Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"bdbf45a8b00766301aa22a8053e713658930ad96bdbce4117be384c383a58708","abstract_canon_sha256":"6975622b6464499ccff448a313a8501b2d6128fdf35b602f3e0da5066f91828d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:23:12.466305Z","signature_b64":"oaBWFDal1tB2cTeUuuW2vrpO8zNgyVoBozhhTCI1yPp76q/o95wUHhBaOZbjl5mcNG7ISB9Io+ukma+z6sfCCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7fb0e86759321c1a8e4d7e55d3cefced1c78c9124ae4c7b46f21915f2aa01ed4","last_reissued_at":"2026-07-05T08:23:12.465831Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:23:12.465831Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Pre-training with Synthetic Data Helps Offline Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Che Wang, Keith Ross, Zecheng Wang, Zixuan Dong","submitted_at":"2023-10-01T19:32:14Z","abstract_excerpt":"Recently, it has been shown that for offline deep reinforcement learning (DRL), pre-training Decision Transformer with a large language corpus can improve downstream performance (Reid et al., 2022). A natural question to ask is whether this performance gain can only be achieved with language pre-training, or can be achieved with simpler pre-training schemes which do not involve language. In this paper, we first show that language is not essential for improved performance, and indeed pre-training with synthetic IID data for a small number of updates can match the performance gains from pre-trai"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.00771","kind":"arxiv","version":4},"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/2310.00771/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":"2310.00771","created_at":"2026-07-05T08:23:12.465895+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.00771v4","created_at":"2026-07-05T08:23:12.465895+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.00771","created_at":"2026-07-05T08:23:12.465895+00:00"},{"alias_kind":"pith_short_12","alias_value":"P6YOQZ2ZGIOB","created_at":"2026-07-05T08:23:12.465895+00:00"},{"alias_kind":"pith_short_16","alias_value":"P6YOQZ2ZGIOBVDSN","created_at":"2026-07-05T08:23:12.465895+00:00"},{"alias_kind":"pith_short_8","alias_value":"P6YOQZ2Z","created_at":"2026-07-05T08:23:12.465895+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/P6YOQZ2ZGIOBVDSNPZK5HTX45U","json":"https://pith.science/pith/P6YOQZ2ZGIOBVDSNPZK5HTX45U.json","graph_json":"https://pith.science/api/pith-number/P6YOQZ2ZGIOBVDSNPZK5HTX45U/graph.json","events_json":"https://pith.science/api/pith-number/P6YOQZ2ZGIOBVDSNPZK5HTX45U/events.json","paper":"https://pith.science/paper/P6YOQZ2Z"},"agent_actions":{"view_html":"https://pith.science/pith/P6YOQZ2ZGIOBVDSNPZK5HTX45U","download_json":"https://pith.science/pith/P6YOQZ2ZGIOBVDSNPZK5HTX45U.json","view_paper":"https://pith.science/paper/P6YOQZ2Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.00771&json=true","fetch_graph":"https://pith.science/api/pith-number/P6YOQZ2ZGIOBVDSNPZK5HTX45U/graph.json","fetch_events":"https://pith.science/api/pith-number/P6YOQZ2ZGIOBVDSNPZK5HTX45U/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P6YOQZ2ZGIOBVDSNPZK5HTX45U/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P6YOQZ2ZGIOBVDSNPZK5HTX45U/action/storage_attestation","attest_author":"https://pith.science/pith/P6YOQZ2ZGIOBVDSNPZK5HTX45U/action/author_attestation","sign_citation":"https://pith.science/pith/P6YOQZ2ZGIOBVDSNPZK5HTX45U/action/citation_signature","submit_replication":"https://pith.science/pith/P6YOQZ2ZGIOBVDSNPZK5HTX45U/action/replication_record"}},"created_at":"2026-07-05T08:23:12.465895+00:00","updated_at":"2026-07-05T08:23:12.465895+00:00"}