{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DI6LSVBK6EQ4EBYNDIDENOIXFX","short_pith_number":"pith:DI6LSVBK","schema_version":"1.0","canonical_sha256":"1a3cb9542af121c2070d1a0646b9172dca7971015450b39487dad57e9289f4cd","source":{"kind":"arxiv","id":"2606.03094","version":1},"attestation_state":"computed","paper":{"title":"FGRPO: Federated GRPO with Adaptive Aggregation on Non-IID Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Feng Li, Jun Luo, Kai Han, Kai Wang, Pengyu Chen, Shaowei Li, Yunsheng Yuan","submitted_at":"2026-06-02T03:32:32Z","abstract_excerpt":"Recent advances in language models have established reinforcement learning as the primary paradigm for eliciting self-correction and long-chain reasoning. While group relative policy optimization (GRPO) offers superior scalability by eliminating the critic network, deploying it on a central infrastructure entails collecting a large volume of data from distributed owners, which poses significant privacy risks. To address these concerns, we introduce federated GRPO (FGRPO), a framework designed to decentralize the fine-tuning of reasoning models across heterogeneous data owners. To effectively m"},"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":"2606.03094","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-06-02T03:32:32Z","cross_cats_sorted":[],"title_canon_sha256":"0a89817ca8dcc5f460009426e0c5f1b1d168c275f32215c65ad81693d18afe42","abstract_canon_sha256":"b1f4578394f74d785d83a93d6c4e50c2cdabc2038a0f0f4fb2f55fefb47452cf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-03T01:05:31.493185Z","signature_b64":"z/RFLInI4Y84aX6keY+3ICFZpVgdinCmfO1NP6jP9d+mHnxcivjnsBNorD8Zcf54cbOB7RHp0fx2z7qD0WfcAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1a3cb9542af121c2070d1a0646b9172dca7971015450b39487dad57e9289f4cd","last_reissued_at":"2026-06-03T01:05:31.492780Z","signature_status":"signed_v1","first_computed_at":"2026-06-03T01:05:31.492780Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FGRPO: Federated GRPO with Adaptive Aggregation on Non-IID Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Feng Li, Jun Luo, Kai Han, Kai Wang, Pengyu Chen, Shaowei Li, Yunsheng Yuan","submitted_at":"2026-06-02T03:32:32Z","abstract_excerpt":"Recent advances in language models have established reinforcement learning as the primary paradigm for eliciting self-correction and long-chain reasoning. While group relative policy optimization (GRPO) offers superior scalability by eliminating the critic network, deploying it on a central infrastructure entails collecting a large volume of data from distributed owners, which poses significant privacy risks. To address these concerns, we introduce federated GRPO (FGRPO), a framework designed to decentralize the fine-tuning of reasoning models across heterogeneous data owners. To effectively m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.03094","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/2606.03094/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":"2606.03094","created_at":"2026-06-03T01:05:31.492828+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.03094v1","created_at":"2026-06-03T01:05:31.492828+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.03094","created_at":"2026-06-03T01:05:31.492828+00:00"},{"alias_kind":"pith_short_12","alias_value":"DI6LSVBK6EQ4","created_at":"2026-06-03T01:05:31.492828+00:00"},{"alias_kind":"pith_short_16","alias_value":"DI6LSVBK6EQ4EBYN","created_at":"2026-06-03T01:05:31.492828+00:00"},{"alias_kind":"pith_short_8","alias_value":"DI6LSVBK","created_at":"2026-06-03T01:05:31.492828+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/DI6LSVBK6EQ4EBYNDIDENOIXFX","json":"https://pith.science/pith/DI6LSVBK6EQ4EBYNDIDENOIXFX.json","graph_json":"https://pith.science/api/pith-number/DI6LSVBK6EQ4EBYNDIDENOIXFX/graph.json","events_json":"https://pith.science/api/pith-number/DI6LSVBK6EQ4EBYNDIDENOIXFX/events.json","paper":"https://pith.science/paper/DI6LSVBK"},"agent_actions":{"view_html":"https://pith.science/pith/DI6LSVBK6EQ4EBYNDIDENOIXFX","download_json":"https://pith.science/pith/DI6LSVBK6EQ4EBYNDIDENOIXFX.json","view_paper":"https://pith.science/paper/DI6LSVBK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.03094&json=true","fetch_graph":"https://pith.science/api/pith-number/DI6LSVBK6EQ4EBYNDIDENOIXFX/graph.json","fetch_events":"https://pith.science/api/pith-number/DI6LSVBK6EQ4EBYNDIDENOIXFX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DI6LSVBK6EQ4EBYNDIDENOIXFX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DI6LSVBK6EQ4EBYNDIDENOIXFX/action/storage_attestation","attest_author":"https://pith.science/pith/DI6LSVBK6EQ4EBYNDIDENOIXFX/action/author_attestation","sign_citation":"https://pith.science/pith/DI6LSVBK6EQ4EBYNDIDENOIXFX/action/citation_signature","submit_replication":"https://pith.science/pith/DI6LSVBK6EQ4EBYNDIDENOIXFX/action/replication_record"}},"created_at":"2026-06-03T01:05:31.492828+00:00","updated_at":"2026-06-03T01:05:31.492828+00:00"}