{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LNP6CGFU64BNKVLKOQBJUBA4FO","short_pith_number":"pith:LNP6CGFU","schema_version":"1.0","canonical_sha256":"5b5fe118b4f702d5556a74029a041c2ba008209f76455d68fa832986a7b8df95","source":{"kind":"arxiv","id":"2606.03017","version":1},"attestation_state":"computed","paper":{"title":"ConTraIRL: Factorized Contrastive Abstractions for Transferable IRL","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Bikramjit Banerjee, Prashant Doshi, Yikang Gui","submitted_at":"2026-06-02T01:47:19Z","abstract_excerpt":"Reward transfer in Inverse Reinforcement Learning (IRL) is unreliable when policies must generalize to unseen combinations of environment dynamics and task goals. We propose Factorized Contrastive Abstractions for Transferable IRL (ConTraIRL), a framework that enables compositional reward transfer by learning decoupled latent representations of these two factors. ConTraIRL uses a dual-encoder architecture that maps observations into separate dynamics and goal latent spaces, trained with a dual contrastive objective. Temporal alignment encourages the dynamics encoder to learn goal-invariant str"},"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":"2606.03017","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-02T01:47:19Z","cross_cats_sorted":["cs.AI","cs.RO"],"title_canon_sha256":"7adf04774a1ad6bcb5599b216bb01bf5da0da560d53199c3f5d390386a9c3a88","abstract_canon_sha256":"43bf24c59d5b02b3eb5b53e2b17ef824c26bf72faae941aafea3bbc0e76da921"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-03T01:05:29.359182Z","signature_b64":"qWZuuAR3JcasmETm+gKY8a/z+Ybuo9DAKeNAeRKsYlAYTtYci39jzLvZkioxbpENiOIXrKTrILLaunXBQu/EDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5b5fe118b4f702d5556a74029a041c2ba008209f76455d68fa832986a7b8df95","last_reissued_at":"2026-06-03T01:05:29.358714Z","signature_status":"signed_v1","first_computed_at":"2026-06-03T01:05:29.358714Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ConTraIRL: Factorized Contrastive Abstractions for Transferable IRL","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Bikramjit Banerjee, Prashant Doshi, Yikang Gui","submitted_at":"2026-06-02T01:47:19Z","abstract_excerpt":"Reward transfer in Inverse Reinforcement Learning (IRL) is unreliable when policies must generalize to unseen combinations of environment dynamics and task goals. We propose Factorized Contrastive Abstractions for Transferable IRL (ConTraIRL), a framework that enables compositional reward transfer by learning decoupled latent representations of these two factors. ConTraIRL uses a dual-encoder architecture that maps observations into separate dynamics and goal latent spaces, trained with a dual contrastive objective. Temporal alignment encourages the dynamics encoder to learn goal-invariant str"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.03017","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/2606.03017/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":"2606.03017","created_at":"2026-06-03T01:05:29.358786+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.03017v1","created_at":"2026-06-03T01:05:29.358786+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.03017","created_at":"2026-06-03T01:05:29.358786+00:00"},{"alias_kind":"pith_short_12","alias_value":"LNP6CGFU64BN","created_at":"2026-06-03T01:05:29.358786+00:00"},{"alias_kind":"pith_short_16","alias_value":"LNP6CGFU64BNKVLK","created_at":"2026-06-03T01:05:29.358786+00:00"},{"alias_kind":"pith_short_8","alias_value":"LNP6CGFU","created_at":"2026-06-03T01:05:29.358786+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/LNP6CGFU64BNKVLKOQBJUBA4FO","json":"https://pith.science/pith/LNP6CGFU64BNKVLKOQBJUBA4FO.json","graph_json":"https://pith.science/api/pith-number/LNP6CGFU64BNKVLKOQBJUBA4FO/graph.json","events_json":"https://pith.science/api/pith-number/LNP6CGFU64BNKVLKOQBJUBA4FO/events.json","paper":"https://pith.science/paper/LNP6CGFU"},"agent_actions":{"view_html":"https://pith.science/pith/LNP6CGFU64BNKVLKOQBJUBA4FO","download_json":"https://pith.science/pith/LNP6CGFU64BNKVLKOQBJUBA4FO.json","view_paper":"https://pith.science/paper/LNP6CGFU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.03017&json=true","fetch_graph":"https://pith.science/api/pith-number/LNP6CGFU64BNKVLKOQBJUBA4FO/graph.json","fetch_events":"https://pith.science/api/pith-number/LNP6CGFU64BNKVLKOQBJUBA4FO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LNP6CGFU64BNKVLKOQBJUBA4FO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LNP6CGFU64BNKVLKOQBJUBA4FO/action/storage_attestation","attest_author":"https://pith.science/pith/LNP6CGFU64BNKVLKOQBJUBA4FO/action/author_attestation","sign_citation":"https://pith.science/pith/LNP6CGFU64BNKVLKOQBJUBA4FO/action/citation_signature","submit_replication":"https://pith.science/pith/LNP6CGFU64BNKVLKOQBJUBA4FO/action/replication_record"}},"created_at":"2026-06-03T01:05:29.358786+00:00","updated_at":"2026-06-03T01:05:29.358786+00:00"}