{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:V7AUVCEG6TCMAG7ZOYQVKZA4UX","short_pith_number":"pith:V7AUVCEG","schema_version":"1.0","canonical_sha256":"afc14a8886f4c4c01bf9762155641ca5cc1f7dec419725a9d5b9a1b2b510d0be","source":{"kind":"arxiv","id":"2410.07096","version":9},"attestation_state":"computed","paper":{"title":"Rejecting Hallucinated State Targets during Planning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Doina Precup, Mingde Zhao, Romain Laroche, Tristan Sylvain, Yoshua Bengio","submitted_at":"2024-10-09T17:35:25Z","abstract_excerpt":"In planning processes of computational decision-making agents, generative or predictive models are often used as \"generators\" to propose \"targets\" representing sets of expected or desirable states. Unfortunately, learned models inevitably hallucinate infeasible targets that can cause delusional behaviors and safety concerns. We first investigate the kinds of infeasible targets that generators can hallucinate. Then, we devise a strategy to identify and reject infeasible targets by learning a target feasibility evaluator. To ensure that the evaluator is robust and non-delusional, we adopted a de"},"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":"2410.07096","kind":"arxiv","version":9},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-10-09T17:35:25Z","cross_cats_sorted":[],"title_canon_sha256":"ec0c52bcdab20793adebb34499ceb1ac7494a79fcddb742e080a95900181be7c","abstract_canon_sha256":"fb552aded4a6cd6a7d93ee9ef53e3e1250be4ab3657245758a01f52bbaba1349"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:51:18.063356Z","signature_b64":"TWyjb6/YsTsg3o4rbniYqaEVr9J//r45BcFL0+au4UTozCK4bzfKoSZff2Xt/atZ1KdFkY7GT1T+bqR5ab84Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"afc14a8886f4c4c01bf9762155641ca5cc1f7dec419725a9d5b9a1b2b510d0be","last_reissued_at":"2026-07-05T11:51:18.062856Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:51:18.062856Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Rejecting Hallucinated State Targets during Planning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Doina Precup, Mingde Zhao, Romain Laroche, Tristan Sylvain, Yoshua Bengio","submitted_at":"2024-10-09T17:35:25Z","abstract_excerpt":"In planning processes of computational decision-making agents, generative or predictive models are often used as \"generators\" to propose \"targets\" representing sets of expected or desirable states. Unfortunately, learned models inevitably hallucinate infeasible targets that can cause delusional behaviors and safety concerns. We first investigate the kinds of infeasible targets that generators can hallucinate. Then, we devise a strategy to identify and reject infeasible targets by learning a target feasibility evaluator. To ensure that the evaluator is robust and non-delusional, we adopted a de"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.07096","kind":"arxiv","version":9},"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/2410.07096/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":"2410.07096","created_at":"2026-07-05T11:51:18.062914+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.07096v9","created_at":"2026-07-05T11:51:18.062914+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.07096","created_at":"2026-07-05T11:51:18.062914+00:00"},{"alias_kind":"pith_short_12","alias_value":"V7AUVCEG6TCM","created_at":"2026-07-05T11:51:18.062914+00:00"},{"alias_kind":"pith_short_16","alias_value":"V7AUVCEG6TCMAG7Z","created_at":"2026-07-05T11:51:18.062914+00:00"},{"alias_kind":"pith_short_8","alias_value":"V7AUVCEG","created_at":"2026-07-05T11:51:18.062914+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/V7AUVCEG6TCMAG7ZOYQVKZA4UX","json":"https://pith.science/pith/V7AUVCEG6TCMAG7ZOYQVKZA4UX.json","graph_json":"https://pith.science/api/pith-number/V7AUVCEG6TCMAG7ZOYQVKZA4UX/graph.json","events_json":"https://pith.science/api/pith-number/V7AUVCEG6TCMAG7ZOYQVKZA4UX/events.json","paper":"https://pith.science/paper/V7AUVCEG"},"agent_actions":{"view_html":"https://pith.science/pith/V7AUVCEG6TCMAG7ZOYQVKZA4UX","download_json":"https://pith.science/pith/V7AUVCEG6TCMAG7ZOYQVKZA4UX.json","view_paper":"https://pith.science/paper/V7AUVCEG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.07096&json=true","fetch_graph":"https://pith.science/api/pith-number/V7AUVCEG6TCMAG7ZOYQVKZA4UX/graph.json","fetch_events":"https://pith.science/api/pith-number/V7AUVCEG6TCMAG7ZOYQVKZA4UX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V7AUVCEG6TCMAG7ZOYQVKZA4UX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V7AUVCEG6TCMAG7ZOYQVKZA4UX/action/storage_attestation","attest_author":"https://pith.science/pith/V7AUVCEG6TCMAG7ZOYQVKZA4UX/action/author_attestation","sign_citation":"https://pith.science/pith/V7AUVCEG6TCMAG7ZOYQVKZA4UX/action/citation_signature","submit_replication":"https://pith.science/pith/V7AUVCEG6TCMAG7ZOYQVKZA4UX/action/replication_record"}},"created_at":"2026-07-05T11:51:18.062914+00:00","updated_at":"2026-07-05T11:51:18.062914+00:00"}