{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:364SNZSI6RPCSZEWSDSRABLKRU","short_pith_number":"pith:364SNZSI","schema_version":"1.0","canonical_sha256":"dfb926e648f45e29649690e510056a8d266e07a1f9e046a8191619873c1faa7a","source":{"kind":"arxiv","id":"2510.17059","version":2},"attestation_state":"computed","paper":{"title":"Consistent Zero-Shot Imitation with Contrastive Goal Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Benjamin Eysenbach, Chongyi Zheng, Kathryn Wantlin","submitted_at":"2025-10-20T00:28:03Z","abstract_excerpt":"Zero-shot imitation learning requires an agent to reproduce expert behavior from a single demonstration without additional environment interaction or gradient updates at test time. We introduce Contrastive Inverse Reinforcement Learning (CIRL), a self-supervised framework for pre-training zero-shot imitation agents. Our methods rests on a key observation that many useful tasks can be summarized by a single goal state. We can thus convert the multi-task inverse RL problem into a more tractable goal-inference problem, and utilize state-of-the-art goal-conditioned RL methods to recover a policy t"},"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":"2510.17059","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-10-20T00:28:03Z","cross_cats_sorted":[],"title_canon_sha256":"27a1e6c4be930e9b075cdb878fe3eeb81df8366c77af2864b2b65d5902d78001","abstract_canon_sha256":"d8b3ba9bbf6aa211186836d2e7e236215a686a12a55e02112887f417661c924c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-25T01:17:47.000607Z","signature_b64":"DhCDSBew4rMYQfvo1oMqXNRYzdEXpKgCDqwefOYm5M8n21iVofumZnlIijT7rUPO5cE683/+Xs1hZe4ss9oLAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dfb926e648f45e29649690e510056a8d266e07a1f9e046a8191619873c1faa7a","last_reissued_at":"2026-06-25T01:17:47.000150Z","signature_status":"signed_v1","first_computed_at":"2026-06-25T01:17:47.000150Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Consistent Zero-Shot Imitation with Contrastive Goal Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Benjamin Eysenbach, Chongyi Zheng, Kathryn Wantlin","submitted_at":"2025-10-20T00:28:03Z","abstract_excerpt":"Zero-shot imitation learning requires an agent to reproduce expert behavior from a single demonstration without additional environment interaction or gradient updates at test time. We introduce Contrastive Inverse Reinforcement Learning (CIRL), a self-supervised framework for pre-training zero-shot imitation agents. Our methods rests on a key observation that many useful tasks can be summarized by a single goal state. We can thus convert the multi-task inverse RL problem into a more tractable goal-inference problem, and utilize state-of-the-art goal-conditioned RL methods to recover a policy t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.17059","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/2510.17059/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":"2510.17059","created_at":"2026-06-25T01:17:47.000205+00:00"},{"alias_kind":"arxiv_version","alias_value":"2510.17059v2","created_at":"2026-06-25T01:17:47.000205+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.17059","created_at":"2026-06-25T01:17:47.000205+00:00"},{"alias_kind":"pith_short_12","alias_value":"364SNZSI6RPC","created_at":"2026-06-25T01:17:47.000205+00:00"},{"alias_kind":"pith_short_16","alias_value":"364SNZSI6RPCSZEW","created_at":"2026-06-25T01:17:47.000205+00:00"},{"alias_kind":"pith_short_8","alias_value":"364SNZSI","created_at":"2026-06-25T01:17:47.000205+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/364SNZSI6RPCSZEWSDSRABLKRU","json":"https://pith.science/pith/364SNZSI6RPCSZEWSDSRABLKRU.json","graph_json":"https://pith.science/api/pith-number/364SNZSI6RPCSZEWSDSRABLKRU/graph.json","events_json":"https://pith.science/api/pith-number/364SNZSI6RPCSZEWSDSRABLKRU/events.json","paper":"https://pith.science/paper/364SNZSI"},"agent_actions":{"view_html":"https://pith.science/pith/364SNZSI6RPCSZEWSDSRABLKRU","download_json":"https://pith.science/pith/364SNZSI6RPCSZEWSDSRABLKRU.json","view_paper":"https://pith.science/paper/364SNZSI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2510.17059&json=true","fetch_graph":"https://pith.science/api/pith-number/364SNZSI6RPCSZEWSDSRABLKRU/graph.json","fetch_events":"https://pith.science/api/pith-number/364SNZSI6RPCSZEWSDSRABLKRU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/364SNZSI6RPCSZEWSDSRABLKRU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/364SNZSI6RPCSZEWSDSRABLKRU/action/storage_attestation","attest_author":"https://pith.science/pith/364SNZSI6RPCSZEWSDSRABLKRU/action/author_attestation","sign_citation":"https://pith.science/pith/364SNZSI6RPCSZEWSDSRABLKRU/action/citation_signature","submit_replication":"https://pith.science/pith/364SNZSI6RPCSZEWSDSRABLKRU/action/replication_record"}},"created_at":"2026-06-25T01:17:47.000205+00:00","updated_at":"2026-06-25T01:17:47.000205+00:00"}