{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:C7HZCGQ4XG57XBILJLQGM22JGT","short_pith_number":"pith:C7HZCGQ4","schema_version":"1.0","canonical_sha256":"17cf911a1cb9bbfb850b4ae0666b4934d35c5d5e2690eb2ca3861c3bb34aa361","source":{"kind":"arxiv","id":"2602.19327","version":3},"attestation_state":"computed","paper":{"title":"Soft Sequence Policy Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Maksim Kryzhanovskiy, Roman Ischenko, Svetlana Glazyrina","submitted_at":"2026-02-22T20:21:00Z","abstract_excerpt":"A significant portion of recent research on Large Language Model (LLM) alignment focuses on developing new policy optimization methods based on Group Relative Policy Optimization (GRPO). Two prominent directions have emerged: (i) a shift toward sequence-level importance sampling weights that better align with the sequence-level rewards used in many tasks, and (ii) alternatives to the PPO-style clipping that aim to avoid the associated loss of training signal and entropy collapse. We introduce Soft Sequence Policy Optimization, an off-policy reinforcement learning objective that incorporates so"},"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":"2602.19327","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-02-22T20:21:00Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"0efee914fc0befe87bd6e07753c441235a4d74e95c9ff907e40ae0fbf1b73560","abstract_canon_sha256":"34559ac862a9d2dd7f05ea57a841d1b9edb2578eee30801645f2da1dc7e2ab01"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-05T01:15:22.161354Z","signature_b64":"U7anXg/W7CAbb8EGkcd03EvtV5Y7m6OJpYouSG2cwV/3EZvcTet4t6bF/2ngELsnn3k9c9HSVtwsrNCfaelPBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"17cf911a1cb9bbfb850b4ae0666b4934d35c5d5e2690eb2ca3861c3bb34aa361","last_reissued_at":"2026-06-05T01:15:22.160802Z","signature_status":"signed_v1","first_computed_at":"2026-06-05T01:15:22.160802Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Soft Sequence Policy Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Maksim Kryzhanovskiy, Roman Ischenko, Svetlana Glazyrina","submitted_at":"2026-02-22T20:21:00Z","abstract_excerpt":"A significant portion of recent research on Large Language Model (LLM) alignment focuses on developing new policy optimization methods based on Group Relative Policy Optimization (GRPO). Two prominent directions have emerged: (i) a shift toward sequence-level importance sampling weights that better align with the sequence-level rewards used in many tasks, and (ii) alternatives to the PPO-style clipping that aim to avoid the associated loss of training signal and entropy collapse. We introduce Soft Sequence Policy Optimization, an off-policy reinforcement learning objective that incorporates so"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.19327","kind":"arxiv","version":3},"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/2602.19327/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":"2602.19327","created_at":"2026-06-05T01:15:22.160867+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.19327v3","created_at":"2026-06-05T01:15:22.160867+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.19327","created_at":"2026-06-05T01:15:22.160867+00:00"},{"alias_kind":"pith_short_12","alias_value":"C7HZCGQ4XG57","created_at":"2026-06-05T01:15:22.160867+00:00"},{"alias_kind":"pith_short_16","alias_value":"C7HZCGQ4XG57XBIL","created_at":"2026-06-05T01:15:22.160867+00:00"},{"alias_kind":"pith_short_8","alias_value":"C7HZCGQ4","created_at":"2026-06-05T01:15:22.160867+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/C7HZCGQ4XG57XBILJLQGM22JGT","json":"https://pith.science/pith/C7HZCGQ4XG57XBILJLQGM22JGT.json","graph_json":"https://pith.science/api/pith-number/C7HZCGQ4XG57XBILJLQGM22JGT/graph.json","events_json":"https://pith.science/api/pith-number/C7HZCGQ4XG57XBILJLQGM22JGT/events.json","paper":"https://pith.science/paper/C7HZCGQ4"},"agent_actions":{"view_html":"https://pith.science/pith/C7HZCGQ4XG57XBILJLQGM22JGT","download_json":"https://pith.science/pith/C7HZCGQ4XG57XBILJLQGM22JGT.json","view_paper":"https://pith.science/paper/C7HZCGQ4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.19327&json=true","fetch_graph":"https://pith.science/api/pith-number/C7HZCGQ4XG57XBILJLQGM22JGT/graph.json","fetch_events":"https://pith.science/api/pith-number/C7HZCGQ4XG57XBILJLQGM22JGT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/C7HZCGQ4XG57XBILJLQGM22JGT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/C7HZCGQ4XG57XBILJLQGM22JGT/action/storage_attestation","attest_author":"https://pith.science/pith/C7HZCGQ4XG57XBILJLQGM22JGT/action/author_attestation","sign_citation":"https://pith.science/pith/C7HZCGQ4XG57XBILJLQGM22JGT/action/citation_signature","submit_replication":"https://pith.science/pith/C7HZCGQ4XG57XBILJLQGM22JGT/action/replication_record"}},"created_at":"2026-06-05T01:15:22.160867+00:00","updated_at":"2026-06-05T01:15:22.160867+00:00"}