{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:4VQ4HAWWZO25SPSDCXFQBYX6YR","short_pith_number":"pith:4VQ4HAWW","schema_version":"1.0","canonical_sha256":"e561c382d6cbb5d93e4315cb00e2fec446f3617a04d8e7ac5b6da134156e0a0e","source":{"kind":"arxiv","id":"2505.10442","version":1},"attestation_state":"computed","paper":{"title":"IN-RIL: Interleaved Reinforcement and Imitation Learning for Policy Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Ahmadreza Moradipari, Dechen Gao, Hanchu Zhou, Hang Wang, Iman Soltani, Junshan Zhang, Nejib Ammar, Shatadal Mishra","submitted_at":"2025-05-15T16:01:21Z","abstract_excerpt":"Imitation learning (IL) and reinforcement learning (RL) each offer distinct advantages for robotics policy learning: IL provides stable learning from demonstrations, and RL promotes generalization through exploration. While existing robot learning approaches using IL-based pre-training followed by RL-based fine-tuning are promising, this two-step learning paradigm often suffers from instability and poor sample efficiency during the RL fine-tuning phase. In this work, we introduce IN-RIL, INterleaved Reinforcement learning and Imitation Learning, for policy fine-tuning, which periodically injec"},"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":"2505.10442","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-05-15T16:01:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d5636edc6a7cbe6e5e20f05e927811d5bc45ebc5ed78bb176739d14f5933154e","abstract_canon_sha256":"fb61cf8b9e60193ba7b6ea6dca9c40299dcf39e8c95fd4c1ab57a62204bd0ca1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:03:38.419703Z","signature_b64":"jdJX7X8NYjkonkbdCvj6I+1PYV5Tux/209KQnshwzawrDS85XKzZq6yAHF8sXrPYsQctT4p4hQGaNQg4rBdeBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e561c382d6cbb5d93e4315cb00e2fec446f3617a04d8e7ac5b6da134156e0a0e","last_reissued_at":"2026-07-05T11:03:38.419260Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:03:38.419260Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"IN-RIL: Interleaved Reinforcement and Imitation Learning for Policy Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Ahmadreza Moradipari, Dechen Gao, Hanchu Zhou, Hang Wang, Iman Soltani, Junshan Zhang, Nejib Ammar, Shatadal Mishra","submitted_at":"2025-05-15T16:01:21Z","abstract_excerpt":"Imitation learning (IL) and reinforcement learning (RL) each offer distinct advantages for robotics policy learning: IL provides stable learning from demonstrations, and RL promotes generalization through exploration. While existing robot learning approaches using IL-based pre-training followed by RL-based fine-tuning are promising, this two-step learning paradigm often suffers from instability and poor sample efficiency during the RL fine-tuning phase. In this work, we introduce IN-RIL, INterleaved Reinforcement learning and Imitation Learning, for policy fine-tuning, which periodically injec"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.10442","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/2505.10442/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":"2505.10442","created_at":"2026-07-05T11:03:38.419327+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.10442v1","created_at":"2026-07-05T11:03:38.419327+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.10442","created_at":"2026-07-05T11:03:38.419327+00:00"},{"alias_kind":"pith_short_12","alias_value":"4VQ4HAWWZO25","created_at":"2026-07-05T11:03:38.419327+00:00"},{"alias_kind":"pith_short_16","alias_value":"4VQ4HAWWZO25SPSD","created_at":"2026-07-05T11:03:38.419327+00:00"},{"alias_kind":"pith_short_8","alias_value":"4VQ4HAWW","created_at":"2026-07-05T11:03:38.419327+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.30576","citing_title":"Uncertainty-Aware and Temporally Regulated Expert Advice in Reinforcement Learning for Autonomous Driving","ref_index":21,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4VQ4HAWWZO25SPSDCXFQBYX6YR","json":"https://pith.science/pith/4VQ4HAWWZO25SPSDCXFQBYX6YR.json","graph_json":"https://pith.science/api/pith-number/4VQ4HAWWZO25SPSDCXFQBYX6YR/graph.json","events_json":"https://pith.science/api/pith-number/4VQ4HAWWZO25SPSDCXFQBYX6YR/events.json","paper":"https://pith.science/paper/4VQ4HAWW"},"agent_actions":{"view_html":"https://pith.science/pith/4VQ4HAWWZO25SPSDCXFQBYX6YR","download_json":"https://pith.science/pith/4VQ4HAWWZO25SPSDCXFQBYX6YR.json","view_paper":"https://pith.science/paper/4VQ4HAWW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.10442&json=true","fetch_graph":"https://pith.science/api/pith-number/4VQ4HAWWZO25SPSDCXFQBYX6YR/graph.json","fetch_events":"https://pith.science/api/pith-number/4VQ4HAWWZO25SPSDCXFQBYX6YR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4VQ4HAWWZO25SPSDCXFQBYX6YR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4VQ4HAWWZO25SPSDCXFQBYX6YR/action/storage_attestation","attest_author":"https://pith.science/pith/4VQ4HAWWZO25SPSDCXFQBYX6YR/action/author_attestation","sign_citation":"https://pith.science/pith/4VQ4HAWWZO25SPSDCXFQBYX6YR/action/citation_signature","submit_replication":"https://pith.science/pith/4VQ4HAWWZO25SPSDCXFQBYX6YR/action/replication_record"}},"created_at":"2026-07-05T11:03:38.419327+00:00","updated_at":"2026-07-05T11:03:38.419327+00:00"}