{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CYJFYIL4JZE3IVTEA67JK3DNJF","short_pith_number":"pith:CYJFYIL4","schema_version":"1.0","canonical_sha256":"16125c217c4e49b4566407be956c6d49531f260e3ecfbb0d2761d6f021cdddf3","source":{"kind":"arxiv","id":"2605.30873","version":1},"attestation_state":"computed","paper":{"title":"Federated Variational Preference Alignment with Gumbel-Softmax Prior for Personalized User Preferences","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.DC"],"primary_cat":"cs.LG","authors_text":"Hoyoung Kim, Jabin Koo, Jungseul Ok, Minwoo Jang","submitted_at":"2026-05-29T05:52:21Z","abstract_excerpt":"Federated Learning (FL) offers a privacy-preserving pathway for aligning Large Language Models (LLMs); however, existing frameworks typically enforce a monolithic reward model, inevitably averaging out inherently conflicting user preferences (e.g., helpfulness vs. harmlessness). While Variational Preference Learning (VPL) offers a pathway to personalization, adapting it to decentralized settings presents a fundamental challenge: posterior collapse driven by severe local data scarcity and heterogeneity. In this paper, we propose Federated Variational Preference Alignment with Gumbel-Softmax Pri"},"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":"2605.30873","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-29T05:52:21Z","cross_cats_sorted":["cs.AI","cs.DC"],"title_canon_sha256":"d76a977b0ff2c24d4604856b03b68aadf24b732be09f1d9fa9a2b90ec2c244b7","abstract_canon_sha256":"a628f7f63d2afac862af9812dfa086a508888ffb6e6739e964b0d4c4e05fdfae"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:03:22.311489Z","signature_b64":"bKyx1rRNIpEK2JY7R3gzfWXgX7LpKObW7hxhIrnRaf+V3DCwsBgTtFoxoG+vV/SLebBAHSbBs8V6hPvFs/WsBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"16125c217c4e49b4566407be956c6d49531f260e3ecfbb0d2761d6f021cdddf3","last_reissued_at":"2026-06-01T01:03:22.310625Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:03:22.310625Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Federated Variational Preference Alignment with Gumbel-Softmax Prior for Personalized User Preferences","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.DC"],"primary_cat":"cs.LG","authors_text":"Hoyoung Kim, Jabin Koo, Jungseul Ok, Minwoo Jang","submitted_at":"2026-05-29T05:52:21Z","abstract_excerpt":"Federated Learning (FL) offers a privacy-preserving pathway for aligning Large Language Models (LLMs); however, existing frameworks typically enforce a monolithic reward model, inevitably averaging out inherently conflicting user preferences (e.g., helpfulness vs. harmlessness). While Variational Preference Learning (VPL) offers a pathway to personalization, adapting it to decentralized settings presents a fundamental challenge: posterior collapse driven by severe local data scarcity and heterogeneity. In this paper, we propose Federated Variational Preference Alignment with Gumbel-Softmax Pri"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.30873","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/2605.30873/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":"2605.30873","created_at":"2026-06-01T01:03:22.310760+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.30873v1","created_at":"2026-06-01T01:03:22.310760+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.30873","created_at":"2026-06-01T01:03:22.310760+00:00"},{"alias_kind":"pith_short_12","alias_value":"CYJFYIL4JZE3","created_at":"2026-06-01T01:03:22.310760+00:00"},{"alias_kind":"pith_short_16","alias_value":"CYJFYIL4JZE3IVTE","created_at":"2026-06-01T01:03:22.310760+00:00"},{"alias_kind":"pith_short_8","alias_value":"CYJFYIL4","created_at":"2026-06-01T01:03:22.310760+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/CYJFYIL4JZE3IVTEA67JK3DNJF","json":"https://pith.science/pith/CYJFYIL4JZE3IVTEA67JK3DNJF.json","graph_json":"https://pith.science/api/pith-number/CYJFYIL4JZE3IVTEA67JK3DNJF/graph.json","events_json":"https://pith.science/api/pith-number/CYJFYIL4JZE3IVTEA67JK3DNJF/events.json","paper":"https://pith.science/paper/CYJFYIL4"},"agent_actions":{"view_html":"https://pith.science/pith/CYJFYIL4JZE3IVTEA67JK3DNJF","download_json":"https://pith.science/pith/CYJFYIL4JZE3IVTEA67JK3DNJF.json","view_paper":"https://pith.science/paper/CYJFYIL4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.30873&json=true","fetch_graph":"https://pith.science/api/pith-number/CYJFYIL4JZE3IVTEA67JK3DNJF/graph.json","fetch_events":"https://pith.science/api/pith-number/CYJFYIL4JZE3IVTEA67JK3DNJF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CYJFYIL4JZE3IVTEA67JK3DNJF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CYJFYIL4JZE3IVTEA67JK3DNJF/action/storage_attestation","attest_author":"https://pith.science/pith/CYJFYIL4JZE3IVTEA67JK3DNJF/action/author_attestation","sign_citation":"https://pith.science/pith/CYJFYIL4JZE3IVTEA67JK3DNJF/action/citation_signature","submit_replication":"https://pith.science/pith/CYJFYIL4JZE3IVTEA67JK3DNJF/action/replication_record"}},"created_at":"2026-06-01T01:03:22.310760+00:00","updated_at":"2026-06-01T01:03:22.310760+00:00"}