{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:PRJWYROQ37GX5Y4M4FAPFAWSDF","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"772e0d507d20bf4bc6ad9fdc6180b22ca10e59b31301e99564cbbf63ccdb5b9b","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-09T19:09:48Z","title_canon_sha256":"efc59a81d02eb6e53dee2780a0f27515624df39ebc841338aded657dbd54b8fa"},"schema_version":"1.0","source":{"id":"2402.06734","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.06734","created_at":"2026-07-01T00:17:07Z"},{"alias_kind":"arxiv_version","alias_value":"2402.06734v2","created_at":"2026-07-01T00:17:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.06734","created_at":"2026-07-01T00:17:07Z"},{"alias_kind":"pith_short_12","alias_value":"PRJWYROQ37GX","created_at":"2026-07-01T00:17:07Z"},{"alias_kind":"pith_short_16","alias_value":"PRJWYROQ37GX5Y4M","created_at":"2026-07-01T00:17:07Z"},{"alias_kind":"pith_short_8","alias_value":"PRJWYROQ","created_at":"2026-07-01T00:17:07Z"}],"graph_snapshots":[{"event_id":"sha256:b1f372000cd9c4af0e719ed14ffc0b23ab676bf6fb471e69d5a6b55c383d9a34","target":"graph","created_at":"2026-07-01T00:17:07Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2402.06734/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We study data corruption robustness for reinforcement learning with human feedback (RLHF) in an offline setting. Given an offline dataset of pairs of trajectories along with feedback about human preferences, an $\\varepsilon$-fraction of the pairs is corrupted (e.g., feedback flipped or trajectory features manipulated), capturing an adversarial attack or noisy human preferences. We aim to design algorithms that identify a near-optimal policy from the corrupted data, with provable guarantees. Existing theoretical works have separately studied the settings of corruption robust RL (learning from s","authors_text":"Adish Singla, Andi Nika, Debmalya Mandal, Goran Radanovi\\'c, Parameswaran Kamalaruban","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-09T19:09:48Z","title":"Corruption Robust Offline Reinforcement Learning with Human Feedback"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.06734","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:f20f96e2b2e8151bb06e2b849990d130b092f993f191aa66e10a58709e6ed851","target":"record","created_at":"2026-07-01T00:17:07Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"772e0d507d20bf4bc6ad9fdc6180b22ca10e59b31301e99564cbbf63ccdb5b9b","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-09T19:09:48Z","title_canon_sha256":"efc59a81d02eb6e53dee2780a0f27515624df39ebc841338aded657dbd54b8fa"},"schema_version":"1.0","source":{"id":"2402.06734","kind":"arxiv","version":2}},"canonical_sha256":"7c536c45d0dfcd7ee38ce140f282d21952e60bd79a811e91dff8fda9e277c7b2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7c536c45d0dfcd7ee38ce140f282d21952e60bd79a811e91dff8fda9e277c7b2","first_computed_at":"2026-07-01T00:17:07.199558Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-01T00:17:07.199558Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"hS6wP85T01y0jBd2mwhjjHluz1qkNwrUXkSv8MxGWGRQ/NXgByG7G5SgWNaiE0XtXgJ6SLLQHUsW0r1QTq0aCA==","signature_status":"signed_v1","signed_at":"2026-07-01T00:17:07.200106Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.06734","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f20f96e2b2e8151bb06e2b849990d130b092f993f191aa66e10a58709e6ed851","sha256:b1f372000cd9c4af0e719ed14ffc0b23ab676bf6fb471e69d5a6b55c383d9a34"],"state_sha256":"d174197142c4bc3e46013838c013554d3da5a12f08db6e990ccc941a0a92de6d"}