{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GGDGG2D3X6DN2RSL7OLBZFKBEI","short_pith_number":"pith:GGDGG2D3","schema_version":"1.0","canonical_sha256":"318663687bbf86dd464bfb961c9541222aba9b57fbcfde412c5044392681b19c","source":{"kind":"arxiv","id":"2608.00750","version":1},"attestation_state":"computed","paper":{"title":"Hierarchical Residual Policy Optimization for Generative Recommendations","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Fukang Yang, Guolei Zeng, Jingtong Gao, Kaifeng Guo, Peng Jiang, Qingpeng Cai, Xiangyu Zhao, Yiming Yang, Yukang Liang","submitted_at":"2026-08-01T16:33:15Z","abstract_excerpt":"Generative recommenders select items by autoregressively decoding semantic identifiers (SIDs), whose token positions induce a coarse-to-fine hierarchy over the item space. In practice, SID decoders are trained via supervised next-token prediction, which imitates logged trajectories rather than directly optimizing downstream utility. This motivates post-training with outcome feedback to guide decoding toward higher utility. However, logged feedback is only observed for the final exposed item, causing most post-training methods to operate at the item level and broadcast the same terminal signal "},"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":"2608.00750","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.IR","submitted_at":"2026-08-01T16:33:15Z","cross_cats_sorted":[],"title_canon_sha256":"6b89105cb1c62f4894fb1bb13910f97be489323db2012128abcfe4ffc05f0bea","abstract_canon_sha256":"823d6611dc32c6f6b2c98388c4db991c76890b2276a333450ea5ca6266811892"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T01:54:52.281438Z","signature_b64":"/QqDVhlGIzay2x5NuyHkTbgI7nzfIlpuXU0DiRproUaAGyPNuVqWgWdD/a78th3T6CklrIuUIqbnRFbUUznAAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"318663687bbf86dd464bfb961c9541222aba9b57fbcfde412c5044392681b19c","last_reissued_at":"2026-08-04T01:54:52.279852Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T01:54:52.279852Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hierarchical Residual Policy Optimization for Generative Recommendations","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Fukang Yang, Guolei Zeng, Jingtong Gao, Kaifeng Guo, Peng Jiang, Qingpeng Cai, Xiangyu Zhao, Yiming Yang, Yukang Liang","submitted_at":"2026-08-01T16:33:15Z","abstract_excerpt":"Generative recommenders select items by autoregressively decoding semantic identifiers (SIDs), whose token positions induce a coarse-to-fine hierarchy over the item space. In practice, SID decoders are trained via supervised next-token prediction, which imitates logged trajectories rather than directly optimizing downstream utility. This motivates post-training with outcome feedback to guide decoding toward higher utility. However, logged feedback is only observed for the final exposed item, causing most post-training methods to operate at the item level and broadcast the same terminal signal "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.00750","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/2608.00750/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":"2608.00750","created_at":"2026-08-04T01:54:52.281230+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.00750v1","created_at":"2026-08-04T01:54:52.281230+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.00750","created_at":"2026-08-04T01:54:52.281230+00:00"},{"alias_kind":"pith_short_12","alias_value":"GGDGG2D3X6DN","created_at":"2026-08-04T01:54:52.281230+00:00"},{"alias_kind":"pith_short_16","alias_value":"GGDGG2D3X6DN2RSL","created_at":"2026-08-04T01:54:52.281230+00:00"},{"alias_kind":"pith_short_8","alias_value":"GGDGG2D3","created_at":"2026-08-04T01:54:52.281230+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/GGDGG2D3X6DN2RSL7OLBZFKBEI","json":"https://pith.science/pith/GGDGG2D3X6DN2RSL7OLBZFKBEI.json","graph_json":"https://pith.science/api/pith-number/GGDGG2D3X6DN2RSL7OLBZFKBEI/graph.json","events_json":"https://pith.science/api/pith-number/GGDGG2D3X6DN2RSL7OLBZFKBEI/events.json","paper":"https://pith.science/paper/GGDGG2D3"},"agent_actions":{"view_html":"https://pith.science/pith/GGDGG2D3X6DN2RSL7OLBZFKBEI","download_json":"https://pith.science/pith/GGDGG2D3X6DN2RSL7OLBZFKBEI.json","view_paper":"https://pith.science/paper/GGDGG2D3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.00750&json=true","fetch_graph":"https://pith.science/api/pith-number/GGDGG2D3X6DN2RSL7OLBZFKBEI/graph.json","fetch_events":"https://pith.science/api/pith-number/GGDGG2D3X6DN2RSL7OLBZFKBEI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GGDGG2D3X6DN2RSL7OLBZFKBEI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GGDGG2D3X6DN2RSL7OLBZFKBEI/action/storage_attestation","attest_author":"https://pith.science/pith/GGDGG2D3X6DN2RSL7OLBZFKBEI/action/author_attestation","sign_citation":"https://pith.science/pith/GGDGG2D3X6DN2RSL7OLBZFKBEI/action/citation_signature","submit_replication":"https://pith.science/pith/GGDGG2D3X6DN2RSL7OLBZFKBEI/action/replication_record"}},"created_at":"2026-08-04T01:54:52.281230+00:00","updated_at":"2026-08-04T01:54:52.281230+00:00"}