{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:IDBLYARW2K2BK5YRTJGC6F4MW3","short_pith_number":"pith:IDBLYARW","schema_version":"1.0","canonical_sha256":"40c2bc0236d2b41577119a4c2f178cb6dfce34987ea6953edafb6e23edd55883","source":{"kind":"arxiv","id":"2606.00840","version":1},"attestation_state":"computed","paper":{"title":"Certificate-Guided Evaluation of Reinforcement Learning Generalization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"{\\DJ}or{\\dj}e \\v{Z}ikeli\\'c, Suguman Bansal, Vignesh Subramanian","submitted_at":"2026-05-30T18:31:57Z","abstract_excerpt":"This work presents a logic-driven framework to evaluate the performance of reinforcement learning (RL) algorithms in their ability to generalize to unseen tasks. Our framework defines a family of inductive reach-avoid tasks, characterized by structural similarities in task dynamics, enabling evaluation of generalization capabilities. We introduce a neural certificate function that validates trajectories generated by RL algorithms by enforcing key conditions, thereby serving as a litmus test for RL generalization. We empirically demonstrate our method's capability in certifying generalization f"},"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":"2606.00840","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-30T18:31:57Z","cross_cats_sorted":[],"title_canon_sha256":"c7fdc440cfe0afb8db7d9cf8bb92adbb93286e10b68efbb698301c1bf8512d64","abstract_canon_sha256":"7f5725069f6f545bb587a47320d9a80aea0babdf04c1115781fc647f01bb9565"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:04:07.321410Z","signature_b64":"C5BrIsr4LD0fCTulaMimgF99IJ33nfRIG53d4dd5YXtiWVuMvyx+hPOYILk1ri0NsQk8q+nXDZvMp1ifuj8/CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"40c2bc0236d2b41577119a4c2f178cb6dfce34987ea6953edafb6e23edd55883","last_reissued_at":"2026-06-02T01:04:07.321058Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:04:07.321058Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Certificate-Guided Evaluation of Reinforcement Learning Generalization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"{\\DJ}or{\\dj}e \\v{Z}ikeli\\'c, Suguman Bansal, Vignesh Subramanian","submitted_at":"2026-05-30T18:31:57Z","abstract_excerpt":"This work presents a logic-driven framework to evaluate the performance of reinforcement learning (RL) algorithms in their ability to generalize to unseen tasks. Our framework defines a family of inductive reach-avoid tasks, characterized by structural similarities in task dynamics, enabling evaluation of generalization capabilities. We introduce a neural certificate function that validates trajectories generated by RL algorithms by enforcing key conditions, thereby serving as a litmus test for RL generalization. We empirically demonstrate our method's capability in certifying generalization f"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00840","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/2606.00840/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":"2606.00840","created_at":"2026-06-02T01:04:07.321115+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.00840v1","created_at":"2026-06-02T01:04:07.321115+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00840","created_at":"2026-06-02T01:04:07.321115+00:00"},{"alias_kind":"pith_short_12","alias_value":"IDBLYARW2K2B","created_at":"2026-06-02T01:04:07.321115+00:00"},{"alias_kind":"pith_short_16","alias_value":"IDBLYARW2K2BK5YR","created_at":"2026-06-02T01:04:07.321115+00:00"},{"alias_kind":"pith_short_8","alias_value":"IDBLYARW","created_at":"2026-06-02T01:04:07.321115+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/IDBLYARW2K2BK5YRTJGC6F4MW3","json":"https://pith.science/pith/IDBLYARW2K2BK5YRTJGC6F4MW3.json","graph_json":"https://pith.science/api/pith-number/IDBLYARW2K2BK5YRTJGC6F4MW3/graph.json","events_json":"https://pith.science/api/pith-number/IDBLYARW2K2BK5YRTJGC6F4MW3/events.json","paper":"https://pith.science/paper/IDBLYARW"},"agent_actions":{"view_html":"https://pith.science/pith/IDBLYARW2K2BK5YRTJGC6F4MW3","download_json":"https://pith.science/pith/IDBLYARW2K2BK5YRTJGC6F4MW3.json","view_paper":"https://pith.science/paper/IDBLYARW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.00840&json=true","fetch_graph":"https://pith.science/api/pith-number/IDBLYARW2K2BK5YRTJGC6F4MW3/graph.json","fetch_events":"https://pith.science/api/pith-number/IDBLYARW2K2BK5YRTJGC6F4MW3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IDBLYARW2K2BK5YRTJGC6F4MW3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IDBLYARW2K2BK5YRTJGC6F4MW3/action/storage_attestation","attest_author":"https://pith.science/pith/IDBLYARW2K2BK5YRTJGC6F4MW3/action/author_attestation","sign_citation":"https://pith.science/pith/IDBLYARW2K2BK5YRTJGC6F4MW3/action/citation_signature","submit_replication":"https://pith.science/pith/IDBLYARW2K2BK5YRTJGC6F4MW3/action/replication_record"}},"created_at":"2026-06-02T01:04:07.321115+00:00","updated_at":"2026-06-02T01:04:07.321115+00:00"}