{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:6NFQSQCMS5DUAZCQ3KCQ2O3AIT","short_pith_number":"pith:6NFQSQCM","schema_version":"1.0","canonical_sha256":"f34b09404c9747406450da850d3b6044dcf5e42a5f770d13cd4340b79667c306","source":{"kind":"arxiv","id":"2307.06333","version":2},"attestation_state":"computed","paper":{"title":"Diagnosis, Feedback, Adaptation: A Human-in-the-Loop Framework for Test-Time Policy Adaptation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.HC","cs.RO"],"primary_cat":"cs.LG","authors_text":"Andi Peng, Andreea Bobu, Aviv Netanyahu, Julie Shah, Mark Ho, Pulkit Agrawal, Tianmin Shu","submitted_at":"2023-07-12T17:55:08Z","abstract_excerpt":"Policies often fail due to distribution shift -- changes in the state and reward that occur when a policy is deployed in new environments. Data augmentation can increase robustness by making the model invariant to task-irrelevant changes in the agent's observation. However, designers don't know which concepts are irrelevant a priori, especially when different end users have different preferences about how the task is performed. We propose an interactive framework to leverage feedback directly from the user to identify personalized task-irrelevant concepts. Our key idea is to generate counterfa"},"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":"2307.06333","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-07-12T17:55:08Z","cross_cats_sorted":["cs.AI","cs.HC","cs.RO"],"title_canon_sha256":"8cd33c15e8b3beabb4ed941c802fcccf820f3d660311aa76f45ba1b77253cb29","abstract_canon_sha256":"6d36959d80858dea339732a2a33bc6c02e519765afb7c0bea5af7335ad90ba16"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:32:40.625313Z","signature_b64":"SFyuz1OZoLtFZ/LoykQiy9uOej6vehhO5RQUmXfC8MSFyuTBfzQ2OxhfL7PbS5KgxT0Xav5RaFXHEyawc4f/Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f34b09404c9747406450da850d3b6044dcf5e42a5f770d13cd4340b79667c306","last_reissued_at":"2026-07-05T06:32:40.624790Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:32:40.624790Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Diagnosis, Feedback, Adaptation: A Human-in-the-Loop Framework for Test-Time Policy Adaptation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.HC","cs.RO"],"primary_cat":"cs.LG","authors_text":"Andi Peng, Andreea Bobu, Aviv Netanyahu, Julie Shah, Mark Ho, Pulkit Agrawal, Tianmin Shu","submitted_at":"2023-07-12T17:55:08Z","abstract_excerpt":"Policies often fail due to distribution shift -- changes in the state and reward that occur when a policy is deployed in new environments. Data augmentation can increase robustness by making the model invariant to task-irrelevant changes in the agent's observation. However, designers don't know which concepts are irrelevant a priori, especially when different end users have different preferences about how the task is performed. We propose an interactive framework to leverage feedback directly from the user to identify personalized task-irrelevant concepts. Our key idea is to generate counterfa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.06333","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/2307.06333/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":"2307.06333","created_at":"2026-07-05T06:32:40.624856+00:00"},{"alias_kind":"arxiv_version","alias_value":"2307.06333v2","created_at":"2026-07-05T06:32:40.624856+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.06333","created_at":"2026-07-05T06:32:40.624856+00:00"},{"alias_kind":"pith_short_12","alias_value":"6NFQSQCMS5DU","created_at":"2026-07-05T06:32:40.624856+00:00"},{"alias_kind":"pith_short_16","alias_value":"6NFQSQCMS5DUAZCQ","created_at":"2026-07-05T06:32:40.624856+00:00"},{"alias_kind":"pith_short_8","alias_value":"6NFQSQCM","created_at":"2026-07-05T06:32:40.624856+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.19672","citing_title":"Alignment and Safety in Large Language Models: Safety Mechanisms, Training Paradigms, and Emerging Challenges","ref_index":201,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6NFQSQCMS5DUAZCQ3KCQ2O3AIT","json":"https://pith.science/pith/6NFQSQCMS5DUAZCQ3KCQ2O3AIT.json","graph_json":"https://pith.science/api/pith-number/6NFQSQCMS5DUAZCQ3KCQ2O3AIT/graph.json","events_json":"https://pith.science/api/pith-number/6NFQSQCMS5DUAZCQ3KCQ2O3AIT/events.json","paper":"https://pith.science/paper/6NFQSQCM"},"agent_actions":{"view_html":"https://pith.science/pith/6NFQSQCMS5DUAZCQ3KCQ2O3AIT","download_json":"https://pith.science/pith/6NFQSQCMS5DUAZCQ3KCQ2O3AIT.json","view_paper":"https://pith.science/paper/6NFQSQCM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2307.06333&json=true","fetch_graph":"https://pith.science/api/pith-number/6NFQSQCMS5DUAZCQ3KCQ2O3AIT/graph.json","fetch_events":"https://pith.science/api/pith-number/6NFQSQCMS5DUAZCQ3KCQ2O3AIT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6NFQSQCMS5DUAZCQ3KCQ2O3AIT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6NFQSQCMS5DUAZCQ3KCQ2O3AIT/action/storage_attestation","attest_author":"https://pith.science/pith/6NFQSQCMS5DUAZCQ3KCQ2O3AIT/action/author_attestation","sign_citation":"https://pith.science/pith/6NFQSQCMS5DUAZCQ3KCQ2O3AIT/action/citation_signature","submit_replication":"https://pith.science/pith/6NFQSQCMS5DUAZCQ3KCQ2O3AIT/action/replication_record"}},"created_at":"2026-07-05T06:32:40.624856+00:00","updated_at":"2026-07-05T06:32:40.624856+00:00"}