{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ZAPMIWALYOPZ66RWMCXM5I2W5G","short_pith_number":"pith:ZAPMIWAL","schema_version":"1.0","canonical_sha256":"c81ec4580bc39f9f7a3660aecea356e9973b194a9b80f047c57c43596788f6d7","source":{"kind":"arxiv","id":"2606.05152","version":1},"attestation_state":"computed","paper":{"title":"Reinforcement Learning from Rich Feedback with Distributional DAgger","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Jacob Fein-Ashley, Paria Rashidinejad, Rishabh Agrawal","submitted_at":"2026-06-03T17:54:04Z","abstract_excerpt":"Reasoning models have advanced rapidly, but the dominant reinforcement learning from verifiable rewards (RLVR) recipe remains surprisingly narrow: sample many responses and reward each with a single bit indicating whether the final answer is correct. Yet many settings provide rich feedback, including execution traces, tool outputs, expert corrections, and model self-evaluations. We study how to use such feedback through a distributional variant of the classic imitation learning algorithm DAgger, where the learner has local access to an expert distribution on states visited by the current polic"},"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":"2606.05152","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-03T17:54:04Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"16f9dba2dd3bcb13b6b67e2f61130416de9e3f83c968e067f12d7ab623cbf638","abstract_canon_sha256":"131a9f66f49bf099bd275726121e8b6d31e15d46ecf4a6bdaa77545834a00015"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-04T02:07:54.175388Z","signature_b64":"oPNtsBd3txS/Eq+fXKhNp+QqfZ56j86d8fZvsmOyDzO6Q9HioCStL5RJUiplG8Itpb4CqyS/zMjwnheNuxbYDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c81ec4580bc39f9f7a3660aecea356e9973b194a9b80f047c57c43596788f6d7","last_reissued_at":"2026-06-04T02:07:54.174843Z","signature_status":"signed_v1","first_computed_at":"2026-06-04T02:07:54.174843Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reinforcement Learning from Rich Feedback with Distributional DAgger","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Jacob Fein-Ashley, Paria Rashidinejad, Rishabh Agrawal","submitted_at":"2026-06-03T17:54:04Z","abstract_excerpt":"Reasoning models have advanced rapidly, but the dominant reinforcement learning from verifiable rewards (RLVR) recipe remains surprisingly narrow: sample many responses and reward each with a single bit indicating whether the final answer is correct. Yet many settings provide rich feedback, including execution traces, tool outputs, expert corrections, and model self-evaluations. We study how to use such feedback through a distributional variant of the classic imitation learning algorithm DAgger, where the learner has local access to an expert distribution on states visited by the current polic"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.05152","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/2606.05152/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":"2606.05152","created_at":"2026-06-04T02:07:54.174904+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.05152v1","created_at":"2026-06-04T02:07:54.174904+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.05152","created_at":"2026-06-04T02:07:54.174904+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZAPMIWALYOPZ","created_at":"2026-06-04T02:07:54.174904+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZAPMIWALYOPZ66RW","created_at":"2026-06-04T02:07:54.174904+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZAPMIWAL","created_at":"2026-06-04T02:07:54.174904+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/ZAPMIWALYOPZ66RWMCXM5I2W5G","json":"https://pith.science/pith/ZAPMIWALYOPZ66RWMCXM5I2W5G.json","graph_json":"https://pith.science/api/pith-number/ZAPMIWALYOPZ66RWMCXM5I2W5G/graph.json","events_json":"https://pith.science/api/pith-number/ZAPMIWALYOPZ66RWMCXM5I2W5G/events.json","paper":"https://pith.science/paper/ZAPMIWAL"},"agent_actions":{"view_html":"https://pith.science/pith/ZAPMIWALYOPZ66RWMCXM5I2W5G","download_json":"https://pith.science/pith/ZAPMIWALYOPZ66RWMCXM5I2W5G.json","view_paper":"https://pith.science/paper/ZAPMIWAL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.05152&json=true","fetch_graph":"https://pith.science/api/pith-number/ZAPMIWALYOPZ66RWMCXM5I2W5G/graph.json","fetch_events":"https://pith.science/api/pith-number/ZAPMIWALYOPZ66RWMCXM5I2W5G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZAPMIWALYOPZ66RWMCXM5I2W5G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZAPMIWALYOPZ66RWMCXM5I2W5G/action/storage_attestation","attest_author":"https://pith.science/pith/ZAPMIWALYOPZ66RWMCXM5I2W5G/action/author_attestation","sign_citation":"https://pith.science/pith/ZAPMIWALYOPZ66RWMCXM5I2W5G/action/citation_signature","submit_replication":"https://pith.science/pith/ZAPMIWALYOPZ66RWMCXM5I2W5G/action/replication_record"}},"created_at":"2026-06-04T02:07:54.174904+00:00","updated_at":"2026-06-04T02:07:54.174904+00:00"}