{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:7L6TQY4V5HAWPQPEILHB6JW3Y3","short_pith_number":"pith:7L6TQY4V","schema_version":"1.0","canonical_sha256":"fafd386395e9c167c1e442ce1f26dbc6dedf868784f055c625de5914addedd8c","source":{"kind":"arxiv","id":"2608.07719","version":1},"attestation_state":"computed","paper":{"title":"CODS: Iterative Bellman-Residual Data Selection for Reusable Offline Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Abu Sa-Adat Mohamed Moon-Im Al Ahsan, Anuj Sharma, Ibne Farabi Shihab, Md Najmus Swaqeeb, Sanjeda Akter","submitted_at":"2026-08-07T19:11:54Z","abstract_excerpt":"Offline reinforcement learning repeatedly trains policies from a fixed transition pool, making redundant data costly across seeds and hyperparameters, while naive subsampling can remove rare transitions needed for long-horizon credit assignment. We introduce CODS, a critic-guided selector that alternates between fitting an algorithm-matched critic and acquiring high-residual transitions before freezing a reusable subset. Unlike prioritized replay, CODS produces a static artifact; unlike one-shot residual selection, it refreshes scores as the critic changes. At a 10\\% budget, CODS retains 96.6\\"},"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":"2608.07719","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-08-07T19:11:54Z","cross_cats_sorted":[],"title_canon_sha256":"361adad1c1d6e56ec13cafd2322bb543c3145e7c7024d5219c464c423cce5723","abstract_canon_sha256":"b3c1da0c39fa235e5ad6f4d2dcdd876d6a169310c8c9e52fdb15497c1dd82b17"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-11T00:15:46.577502Z","signature_b64":"GCPbOFCddMq++1yu86uZMNWGol9RP47J9iqlNZ1d5g7PyIaecXRHKMChSNimDz1M8ZlqyZvxSZesxSE7i3aHAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fafd386395e9c167c1e442ce1f26dbc6dedf868784f055c625de5914addedd8c","last_reissued_at":"2026-08-11T00:15:46.575268Z","signature_status":"signed_v1","first_computed_at":"2026-08-11T00:15:46.575268Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CODS: Iterative Bellman-Residual Data Selection for Reusable Offline Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Abu Sa-Adat Mohamed Moon-Im Al Ahsan, Anuj Sharma, Ibne Farabi Shihab, Md Najmus Swaqeeb, Sanjeda Akter","submitted_at":"2026-08-07T19:11:54Z","abstract_excerpt":"Offline reinforcement learning repeatedly trains policies from a fixed transition pool, making redundant data costly across seeds and hyperparameters, while naive subsampling can remove rare transitions needed for long-horizon credit assignment. We introduce CODS, a critic-guided selector that alternates between fitting an algorithm-matched critic and acquiring high-residual transitions before freezing a reusable subset. Unlike prioritized replay, CODS produces a static artifact; unlike one-shot residual selection, it refreshes scores as the critic changes. At a 10\\% budget, CODS retains 96.6\\"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.07719","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/2608.07719/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":"2608.07719","created_at":"2026-08-11T00:15:46.575654+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.07719v1","created_at":"2026-08-11T00:15:46.575654+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.07719","created_at":"2026-08-11T00:15:46.575654+00:00"},{"alias_kind":"pith_short_12","alias_value":"7L6TQY4V5HAW","created_at":"2026-08-11T00:15:46.575654+00:00"},{"alias_kind":"pith_short_16","alias_value":"7L6TQY4V5HAWPQPE","created_at":"2026-08-11T00:15:46.575654+00:00"},{"alias_kind":"pith_short_8","alias_value":"7L6TQY4V","created_at":"2026-08-11T00:15:46.575654+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/7L6TQY4V5HAWPQPEILHB6JW3Y3","json":"https://pith.science/pith/7L6TQY4V5HAWPQPEILHB6JW3Y3.json","graph_json":"https://pith.science/api/pith-number/7L6TQY4V5HAWPQPEILHB6JW3Y3/graph.json","events_json":"https://pith.science/api/pith-number/7L6TQY4V5HAWPQPEILHB6JW3Y3/events.json","paper":"https://pith.science/paper/7L6TQY4V"},"agent_actions":{"view_html":"https://pith.science/pith/7L6TQY4V5HAWPQPEILHB6JW3Y3","download_json":"https://pith.science/pith/7L6TQY4V5HAWPQPEILHB6JW3Y3.json","view_paper":"https://pith.science/paper/7L6TQY4V","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.07719&json=true","fetch_graph":"https://pith.science/api/pith-number/7L6TQY4V5HAWPQPEILHB6JW3Y3/graph.json","fetch_events":"https://pith.science/api/pith-number/7L6TQY4V5HAWPQPEILHB6JW3Y3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7L6TQY4V5HAWPQPEILHB6JW3Y3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7L6TQY4V5HAWPQPEILHB6JW3Y3/action/storage_attestation","attest_author":"https://pith.science/pith/7L6TQY4V5HAWPQPEILHB6JW3Y3/action/author_attestation","sign_citation":"https://pith.science/pith/7L6TQY4V5HAWPQPEILHB6JW3Y3/action/citation_signature","submit_replication":"https://pith.science/pith/7L6TQY4V5HAWPQPEILHB6JW3Y3/action/replication_record"}},"created_at":"2026-08-11T00:15:46.575654+00:00","updated_at":"2026-08-11T00:15:46.575654+00:00"}