{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:WJVZRSFMT3XTEEXAFUOFKOL3RX","short_pith_number":"pith:WJVZRSFM","canonical_record":{"source":{"id":"2505.21182","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-27T13:33:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3508571cd1668d04698edc8506d1808c961d5ef382121eef46f1815faaee45fe","abstract_canon_sha256":"da38b8c713cd9bcd67230e3a8517c9f120e21773a7fba3883e34a7a2c64d731b"},"schema_version":"1.0"},"canonical_sha256":"b26b98c8ac9eef3212e02d1c55397b8ddc2c3830af499e4e5117171448bf010c","source":{"kind":"arxiv","id":"2505.21182","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.21182","created_at":"2026-07-05T11:10:25Z"},{"alias_kind":"arxiv_version","alias_value":"2505.21182v1","created_at":"2026-07-05T11:10:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.21182","created_at":"2026-07-05T11:10:25Z"},{"alias_kind":"pith_short_12","alias_value":"WJVZRSFMT3XT","created_at":"2026-07-05T11:10:25Z"},{"alias_kind":"pith_short_16","alias_value":"WJVZRSFMT3XTEEXA","created_at":"2026-07-05T11:10:25Z"},{"alias_kind":"pith_short_8","alias_value":"WJVZRSFM","created_at":"2026-07-05T11:10:25Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:WJVZRSFMT3XTEEXAFUOFKOL3RX","target":"record","payload":{"canonical_record":{"source":{"id":"2505.21182","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-27T13:33:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3508571cd1668d04698edc8506d1808c961d5ef382121eef46f1815faaee45fe","abstract_canon_sha256":"da38b8c713cd9bcd67230e3a8517c9f120e21773a7fba3883e34a7a2c64d731b"},"schema_version":"1.0"},"canonical_sha256":"b26b98c8ac9eef3212e02d1c55397b8ddc2c3830af499e4e5117171448bf010c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:10:25.150489Z","signature_b64":"JnQMe7bZcuqGyKkgX/BlUhTKCVwzLilQdcnK1tqtvw/OBhV66KMXN1UjahPHH2tSaSAS7aUIrn8t9iEn9L+UAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b26b98c8ac9eef3212e02d1c55397b8ddc2c3830af499e4e5117171448bf010c","last_reissued_at":"2026-07-05T11:10:25.150003Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:10:25.150003Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.21182","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:10:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dcyEjp7kIvgXwBD/vwc7+OeyXFSVI9lftFNJMITBBIRYaly3OexF2VrYqyQbfTDnYQk3Re/CS6evbhuyTw3PBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T13:49:21.495287Z"},"content_sha256":"a1dd2ca05662c23dab09eb23246effc3f0a29852f91a969462638e6fd560371c","schema_version":"1.0","event_id":"sha256:a1dd2ca05662c23dab09eb23246effc3f0a29852f91a969462638e6fd560371c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:WJVZRSFMT3XTEEXAFUOFKOL3RX","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Learning What to Do and What Not To Do: Offline Imitation from Expert and Undesirable Demonstrations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Huy Hoang, Pradeep Varakantham, Tanvi Verma, Tien Mai","submitted_at":"2025-05-27T13:33:21Z","abstract_excerpt":"Offline imitation learning typically learns from expert and unlabeled demonstrations, yet often overlooks the valuable signal in explicitly undesirable behaviors. In this work, we study offline imitation learning from contrasting behaviors, where the dataset contains both expert and undesirable demonstrations. We propose a novel formulation that optimizes a difference of KL divergences over the state-action visitation distributions of expert and undesirable (or bad) data. Although the resulting objective is a DC (Difference-of-Convex) program, we prove that it becomes convex when expert demons"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.21182","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/2505.21182/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:10:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"L6+DF3vIbpYVl9ZaEhpqei/G9F0EBKKkf8D+GawOHMpcVB+mtkBtAalViP355Hn1IblKCPsYtNOkQz7cqZadCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T13:49:21.495676Z"},"content_sha256":"62a35dab9f49cc7a692016605dbc9854df0b775d78a5cee557766e26d19d9ff4","schema_version":"1.0","event_id":"sha256:62a35dab9f49cc7a692016605dbc9854df0b775d78a5cee557766e26d19d9ff4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WJVZRSFMT3XTEEXAFUOFKOL3RX/bundle.json","state_url":"https://pith.science/pith/WJVZRSFMT3XTEEXAFUOFKOL3RX/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WJVZRSFMT3XTEEXAFUOFKOL3RX/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-06T13:49:21Z","links":{"resolver":"https://pith.science/pith/WJVZRSFMT3XTEEXAFUOFKOL3RX","bundle":"https://pith.science/pith/WJVZRSFMT3XTEEXAFUOFKOL3RX/bundle.json","state":"https://pith.science/pith/WJVZRSFMT3XTEEXAFUOFKOL3RX/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WJVZRSFMT3XTEEXAFUOFKOL3RX/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:WJVZRSFMT3XTEEXAFUOFKOL3RX","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":"da38b8c713cd9bcd67230e3a8517c9f120e21773a7fba3883e34a7a2c64d731b","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-27T13:33:21Z","title_canon_sha256":"3508571cd1668d04698edc8506d1808c961d5ef382121eef46f1815faaee45fe"},"schema_version":"1.0","source":{"id":"2505.21182","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.21182","created_at":"2026-07-05T11:10:25Z"},{"alias_kind":"arxiv_version","alias_value":"2505.21182v1","created_at":"2026-07-05T11:10:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.21182","created_at":"2026-07-05T11:10:25Z"},{"alias_kind":"pith_short_12","alias_value":"WJVZRSFMT3XT","created_at":"2026-07-05T11:10:25Z"},{"alias_kind":"pith_short_16","alias_value":"WJVZRSFMT3XTEEXA","created_at":"2026-07-05T11:10:25Z"},{"alias_kind":"pith_short_8","alias_value":"WJVZRSFM","created_at":"2026-07-05T11:10:25Z"}],"graph_snapshots":[{"event_id":"sha256:62a35dab9f49cc7a692016605dbc9854df0b775d78a5cee557766e26d19d9ff4","target":"graph","created_at":"2026-07-05T11:10:25Z","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/2505.21182/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Offline imitation learning typically learns from expert and unlabeled demonstrations, yet often overlooks the valuable signal in explicitly undesirable behaviors. In this work, we study offline imitation learning from contrasting behaviors, where the dataset contains both expert and undesirable demonstrations. We propose a novel formulation that optimizes a difference of KL divergences over the state-action visitation distributions of expert and undesirable (or bad) data. Although the resulting objective is a DC (Difference-of-Convex) program, we prove that it becomes convex when expert demons","authors_text":"Huy Hoang, Pradeep Varakantham, Tanvi Verma, Tien Mai","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-27T13:33:21Z","title":"Learning What to Do and What Not To Do: Offline Imitation from Expert and Undesirable Demonstrations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.21182","kind":"arxiv","version":1},"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:a1dd2ca05662c23dab09eb23246effc3f0a29852f91a969462638e6fd560371c","target":"record","created_at":"2026-07-05T11:10:25Z","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":"da38b8c713cd9bcd67230e3a8517c9f120e21773a7fba3883e34a7a2c64d731b","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-27T13:33:21Z","title_canon_sha256":"3508571cd1668d04698edc8506d1808c961d5ef382121eef46f1815faaee45fe"},"schema_version":"1.0","source":{"id":"2505.21182","kind":"arxiv","version":1}},"canonical_sha256":"b26b98c8ac9eef3212e02d1c55397b8ddc2c3830af499e4e5117171448bf010c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b26b98c8ac9eef3212e02d1c55397b8ddc2c3830af499e4e5117171448bf010c","first_computed_at":"2026-07-05T11:10:25.150003Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:10:25.150003Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"JnQMe7bZcuqGyKkgX/BlUhTKCVwzLilQdcnK1tqtvw/OBhV66KMXN1UjahPHH2tSaSAS7aUIrn8t9iEn9L+UAA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:10:25.150489Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.21182","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a1dd2ca05662c23dab09eb23246effc3f0a29852f91a969462638e6fd560371c","sha256:62a35dab9f49cc7a692016605dbc9854df0b775d78a5cee557766e26d19d9ff4"],"state_sha256":"750ef178183b750f50c1b893a8e58898c04a82d3d7c121a49b21278075d576b6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FqV2a+CzsE6HTck/LnN7Z5UocYolSso6gLlr3+1pLuXTrqzUG8hKj1uEmDD4rQozvA/abhGXtjgdFTDibEOyCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T13:49:21.497660Z","bundle_sha256":"45a24a548848169d43970d1b4c4d6ba6fdbb8bf60b4dace69f758e31b696dbda"}}