{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:UQBBR3MVAF5G7MDW3IRXYOWWN7","short_pith_number":"pith:UQBBR3MV","schema_version":"1.0","canonical_sha256":"a40218ed95017a6fb076da237c3ad66fefa9d6448efb7cce631dfd11dda018ac","source":{"kind":"arxiv","id":"2605.28293","version":1},"attestation_state":"computed","paper":{"title":"ProRL: Effective Reinforcement Learning for Proactive Recommendation via Rectified Policy Gradient Estimation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ao Xu, Denghui Geng, Deqing Yang, Hengrui Chen, Hongru Hou, Jiaqing Liang, Jinhui Huang, Tiehua Mei","submitted_at":"2026-05-27T10:43:37Z","abstract_excerpt":"Proactive Recommender Systems (PRSs) aim to guide user preference shift toward target items by generating paths of intermediate recommendations. Reinforcement learning (RL) provides a principled framework for optimizing such sequential decision tasks, as path rewards can naturally capture both short-term acceptance and long-term guidance effectiveness. However, naively applying policy gradients to PRS results in deficient gradient estimation. We identify two deficiencies: (1) path-level rewards decompose into step-level rewards with positive mean, creating a length-dependent bias that causes g"},"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":"2605.28293","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-27T10:43:37Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"52f222827534008a5bdaeb95a89d9dee61e9453a1b50d9cd6e0504ce506cd50d","abstract_canon_sha256":"92c8e14c67c11c9a3f858f810f1c06fe6d844d10f1831ab3c661c414df261c88"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:05:05.454250Z","signature_b64":"t/1d8IFA2/eVfbGHShnXbuWU1JSXq51rISX24mADTl4eX1W7OAiMpoyz/HJFeGSBNFDepWMWjbNiD+MUEaVWAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a40218ed95017a6fb076da237c3ad66fefa9d6448efb7cce631dfd11dda018ac","last_reissued_at":"2026-05-28T01:05:05.453826Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:05:05.453826Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ProRL: Effective Reinforcement Learning for Proactive Recommendation via Rectified Policy Gradient Estimation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ao Xu, Denghui Geng, Deqing Yang, Hengrui Chen, Hongru Hou, Jiaqing Liang, Jinhui Huang, Tiehua Mei","submitted_at":"2026-05-27T10:43:37Z","abstract_excerpt":"Proactive Recommender Systems (PRSs) aim to guide user preference shift toward target items by generating paths of intermediate recommendations. Reinforcement learning (RL) provides a principled framework for optimizing such sequential decision tasks, as path rewards can naturally capture both short-term acceptance and long-term guidance effectiveness. However, naively applying policy gradients to PRS results in deficient gradient estimation. We identify two deficiencies: (1) path-level rewards decompose into step-level rewards with positive mean, creating a length-dependent bias that causes g"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.28293","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/2605.28293/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":"2605.28293","created_at":"2026-05-28T01:05:05.453892+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.28293v1","created_at":"2026-05-28T01:05:05.453892+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.28293","created_at":"2026-05-28T01:05:05.453892+00:00"},{"alias_kind":"pith_short_12","alias_value":"UQBBR3MVAF5G","created_at":"2026-05-28T01:05:05.453892+00:00"},{"alias_kind":"pith_short_16","alias_value":"UQBBR3MVAF5G7MDW","created_at":"2026-05-28T01:05:05.453892+00:00"},{"alias_kind":"pith_short_8","alias_value":"UQBBR3MV","created_at":"2026-05-28T01:05:05.453892+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/UQBBR3MVAF5G7MDW3IRXYOWWN7","json":"https://pith.science/pith/UQBBR3MVAF5G7MDW3IRXYOWWN7.json","graph_json":"https://pith.science/api/pith-number/UQBBR3MVAF5G7MDW3IRXYOWWN7/graph.json","events_json":"https://pith.science/api/pith-number/UQBBR3MVAF5G7MDW3IRXYOWWN7/events.json","paper":"https://pith.science/paper/UQBBR3MV"},"agent_actions":{"view_html":"https://pith.science/pith/UQBBR3MVAF5G7MDW3IRXYOWWN7","download_json":"https://pith.science/pith/UQBBR3MVAF5G7MDW3IRXYOWWN7.json","view_paper":"https://pith.science/paper/UQBBR3MV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.28293&json=true","fetch_graph":"https://pith.science/api/pith-number/UQBBR3MVAF5G7MDW3IRXYOWWN7/graph.json","fetch_events":"https://pith.science/api/pith-number/UQBBR3MVAF5G7MDW3IRXYOWWN7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UQBBR3MVAF5G7MDW3IRXYOWWN7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UQBBR3MVAF5G7MDW3IRXYOWWN7/action/storage_attestation","attest_author":"https://pith.science/pith/UQBBR3MVAF5G7MDW3IRXYOWWN7/action/author_attestation","sign_citation":"https://pith.science/pith/UQBBR3MVAF5G7MDW3IRXYOWWN7/action/citation_signature","submit_replication":"https://pith.science/pith/UQBBR3MVAF5G7MDW3IRXYOWWN7/action/replication_record"}},"created_at":"2026-05-28T01:05:05.453892+00:00","updated_at":"2026-05-28T01:05:05.453892+00:00"}