{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:C3NWHLP7UCUYPUDA6KXGHSXBIQ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"a1043a179a98c9a4de85d25ca30b7fa04540e7f98ff61de568bf5609318aa03d","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-05T09:13:37Z","title_canon_sha256":"ff995b34a4a2294f7d2e2808adb86553a23ca17d150dfe000f3bdc2c87371681"},"schema_version":"1.0","source":{"id":"2506.05433","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.05433","created_at":"2026-07-05T11:17:02Z"},{"alias_kind":"arxiv_version","alias_value":"2506.05433v1","created_at":"2026-07-05T11:17:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.05433","created_at":"2026-07-05T11:17:02Z"},{"alias_kind":"pith_short_12","alias_value":"C3NWHLP7UCUY","created_at":"2026-07-05T11:17:02Z"},{"alias_kind":"pith_short_16","alias_value":"C3NWHLP7UCUYPUDA","created_at":"2026-07-05T11:17:02Z"},{"alias_kind":"pith_short_8","alias_value":"C3NWHLP7","created_at":"2026-07-05T11:17:02Z"}],"graph_snapshots":[{"event_id":"sha256:dd28e703f0d6b745639001bf5f1c2cd09816d599843bc6be6e615fd2b7a4bf7d","target":"graph","created_at":"2026-07-05T11:17:02Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2506.05433/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Group Relative Policy Optimization (GRPO) enhances policy learning by computing gradients from relative comparisons among candidate outputs that share a common input prefix. Despite its effectiveness, GRPO introduces substantial computational overhead when processing long shared prefixes, which must be redundantly encoded for each group member. This inefficiency becomes a major scalability bottleneck in long-context learning scenarios. We propose Prefix Grouper, an efficient GRPO training algorithm that eliminates redundant prefix computation via a Shared-Prefix Forward strategy. In particular","authors_text":"Jing Liu, Junxian Cai, Longteng Guo, Qingbin Liu, Tongtian Yue, Xi Chen, Yepeng Tang, Zikang Liu","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-05T09:13:37Z","title":"Prefix Grouper: Efficient GRPO Training through Shared-Prefix Forward"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.05433","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:b6196d9c1687bf66a5dcf7fe32f994e8b3d0b37340ac6c26c9791566334e5972","target":"record","created_at":"2026-07-05T11:17:02Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"a1043a179a98c9a4de85d25ca30b7fa04540e7f98ff61de568bf5609318aa03d","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-05T09:13:37Z","title_canon_sha256":"ff995b34a4a2294f7d2e2808adb86553a23ca17d150dfe000f3bdc2c87371681"},"schema_version":"1.0","source":{"id":"2506.05433","kind":"arxiv","version":1}},"canonical_sha256":"16db63adffa0a987d060f2ae63cae1440d02ebe7bc6b2e52419340d0657eb544","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"16db63adffa0a987d060f2ae63cae1440d02ebe7bc6b2e52419340d0657eb544","first_computed_at":"2026-07-05T11:17:02.597500Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:17:02.597500Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+LQM7MLCCN8Di1EB8ClDL0CqSzTehpwbZ+wXKHUYBaJxR46IxClNKYD+DsMPQq4rtQn/TBamYo7nZIkzjSA7AQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:17:02.597978Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.05433","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b6196d9c1687bf66a5dcf7fe32f994e8b3d0b37340ac6c26c9791566334e5972","sha256:dd28e703f0d6b745639001bf5f1c2cd09816d599843bc6be6e615fd2b7a4bf7d"],"state_sha256":"189a12288205ebcea266937b10824787f344dd9a453ed79b4afa1bac93b136f4"}