{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UITG3DLHTTG4IRMOTABG3NU7YA","short_pith_number":"pith:UITG3DLH","schema_version":"1.0","canonical_sha256":"a2266d8d679ccdc4458e98026db69fc0267cd348d5e3311a78fac27722fe192d","source":{"kind":"arxiv","id":"2504.05585","version":2},"attestation_state":"computed","paper":{"title":"TW-CRL: Time-Weighted Contrastive Reward Learning for Efficient Inverse Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ning Yang, Stephen Xia, Yicheng Gao, Yuxuan Li","submitted_at":"2025-04-08T00:48:29Z","abstract_excerpt":"Episodic tasks in Reinforcement Learning (RL) often pose challenges due to sparse reward signals and high-dimensional state spaces, which hinder efficient learning. Additionally, these tasks often feature hidden \"trap states\" -- irreversible failures that prevent task completion but do not provide explicit negative rewards to guide agents away from repeated errors. To address these issues, we propose Time-Weighted Contrastive Reward Learning (TW-CRL), an Inverse Reinforcement Learning (IRL) framework that leverages both successful and failed demonstrations. By incorporating temporal informatio"},"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":"2504.05585","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-08T00:48:29Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"40c6412ca8ae09a72b1f8223148d62ed24a299bd4c477c5b85fcbe5e0804304c","abstract_canon_sha256":"dddee4b9a2d5a6380a607e47ae8242353a8c90c3e0c16ea900f009f64a804888"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:07:23.336833Z","signature_b64":"5NhTlRd9LDSWYGokn8BpuhNTlmqPWmIfn/XvvJl+tvvd9/BCWS9P53wmO4vc+aVbY4qEL32ZW5dlXauWSaWHBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a2266d8d679ccdc4458e98026db69fc0267cd348d5e3311a78fac27722fe192d","last_reissued_at":"2026-07-05T11:07:23.336346Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:07:23.336346Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TW-CRL: Time-Weighted Contrastive Reward Learning for Efficient Inverse Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ning Yang, Stephen Xia, Yicheng Gao, Yuxuan Li","submitted_at":"2025-04-08T00:48:29Z","abstract_excerpt":"Episodic tasks in Reinforcement Learning (RL) often pose challenges due to sparse reward signals and high-dimensional state spaces, which hinder efficient learning. Additionally, these tasks often feature hidden \"trap states\" -- irreversible failures that prevent task completion but do not provide explicit negative rewards to guide agents away from repeated errors. To address these issues, we propose Time-Weighted Contrastive Reward Learning (TW-CRL), an Inverse Reinforcement Learning (IRL) framework that leverages both successful and failed demonstrations. By incorporating temporal informatio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.05585","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/2504.05585/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":"2504.05585","created_at":"2026-07-05T11:07:23.336401+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.05585v2","created_at":"2026-07-05T11:07:23.336401+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.05585","created_at":"2026-07-05T11:07:23.336401+00:00"},{"alias_kind":"pith_short_12","alias_value":"UITG3DLHTTG4","created_at":"2026-07-05T11:07:23.336401+00:00"},{"alias_kind":"pith_short_16","alias_value":"UITG3DLHTTG4IRMO","created_at":"2026-07-05T11:07:23.336401+00:00"},{"alias_kind":"pith_short_8","alias_value":"UITG3DLH","created_at":"2026-07-05T11:07:23.336401+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.03017","citing_title":"ConTraIRL: Factorized Contrastive Abstractions for Transferable IRL","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UITG3DLHTTG4IRMOTABG3NU7YA","json":"https://pith.science/pith/UITG3DLHTTG4IRMOTABG3NU7YA.json","graph_json":"https://pith.science/api/pith-number/UITG3DLHTTG4IRMOTABG3NU7YA/graph.json","events_json":"https://pith.science/api/pith-number/UITG3DLHTTG4IRMOTABG3NU7YA/events.json","paper":"https://pith.science/paper/UITG3DLH"},"agent_actions":{"view_html":"https://pith.science/pith/UITG3DLHTTG4IRMOTABG3NU7YA","download_json":"https://pith.science/pith/UITG3DLHTTG4IRMOTABG3NU7YA.json","view_paper":"https://pith.science/paper/UITG3DLH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.05585&json=true","fetch_graph":"https://pith.science/api/pith-number/UITG3DLHTTG4IRMOTABG3NU7YA/graph.json","fetch_events":"https://pith.science/api/pith-number/UITG3DLHTTG4IRMOTABG3NU7YA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UITG3DLHTTG4IRMOTABG3NU7YA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UITG3DLHTTG4IRMOTABG3NU7YA/action/storage_attestation","attest_author":"https://pith.science/pith/UITG3DLHTTG4IRMOTABG3NU7YA/action/author_attestation","sign_citation":"https://pith.science/pith/UITG3DLHTTG4IRMOTABG3NU7YA/action/citation_signature","submit_replication":"https://pith.science/pith/UITG3DLHTTG4IRMOTABG3NU7YA/action/replication_record"}},"created_at":"2026-07-05T11:07:23.336401+00:00","updated_at":"2026-07-05T11:07:23.336401+00:00"}