{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:TEDAQYBAS7FFPWGIT5KDTXYX3E","short_pith_number":"pith:TEDAQYBA","schema_version":"1.0","canonical_sha256":"990608602097ca57d8c89f5439df17d928e0eb5267330f923d42f9256ca00fbe","source":{"kind":"arxiv","id":"2508.10028","version":1},"attestation_state":"computed","paper":{"title":"PREF: Reference-Free Evaluation of Personalised Text Generation in LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.HC","cs.LG"],"primary_cat":"cs.CL","authors_text":"Aldo Lipani, Bin Wu, Emine Yilmaz, Hossein A. Rahmani, Jerome Ramos, Xiao Fu","submitted_at":"2025-08-08T14:32:31Z","abstract_excerpt":"Personalised text generation is essential for user-centric information systems, yet most evaluation methods overlook the individuality of users. We introduce \\textbf{PREF}, a \\textbf{P}ersonalised \\textbf{R}eference-free \\textbf{E}valuation \\textbf{F}ramework that jointly measures general output quality and user-specific alignment without requiring gold personalised references. PREF operates in a three-step pipeline: (1) a coverage stage uses a large language model (LLM) to generate a comprehensive, query-specific guideline covering universal criteria such as factuality, coherence, and complet"},"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":"2508.10028","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-08T14:32:31Z","cross_cats_sorted":["cs.AI","cs.HC","cs.LG"],"title_canon_sha256":"fbf1a44a319868ec46df9102cbe5a236c12d1a0e43119df400fddf2c3088e60e","abstract_canon_sha256":"74a77604d11952ec521c4f567380166cf546ded8a5fe326a60e80ffd26095e9c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:53:29.512394Z","signature_b64":"1G0lXY5/o+40ljnjgfjHpwu4+UWyzWCpAL7082Ra0bL7T0EdQ4AaqnlK3zHRP6Uhe/Xc3Et6Mj8ppsjWDbPZBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"990608602097ca57d8c89f5439df17d928e0eb5267330f923d42f9256ca00fbe","last_reissued_at":"2026-07-05T11:53:29.511900Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:53:29.511900Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PREF: Reference-Free Evaluation of Personalised Text Generation in LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.HC","cs.LG"],"primary_cat":"cs.CL","authors_text":"Aldo Lipani, Bin Wu, Emine Yilmaz, Hossein A. Rahmani, Jerome Ramos, Xiao Fu","submitted_at":"2025-08-08T14:32:31Z","abstract_excerpt":"Personalised text generation is essential for user-centric information systems, yet most evaluation methods overlook the individuality of users. We introduce \\textbf{PREF}, a \\textbf{P}ersonalised \\textbf{R}eference-free \\textbf{E}valuation \\textbf{F}ramework that jointly measures general output quality and user-specific alignment without requiring gold personalised references. PREF operates in a three-step pipeline: (1) a coverage stage uses a large language model (LLM) to generate a comprehensive, query-specific guideline covering universal criteria such as factuality, coherence, and complet"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.10028","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/2508.10028/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":"2508.10028","created_at":"2026-07-05T11:53:29.511954+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.10028v1","created_at":"2026-07-05T11:53:29.511954+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.10028","created_at":"2026-07-05T11:53:29.511954+00:00"},{"alias_kind":"pith_short_12","alias_value":"TEDAQYBAS7FF","created_at":"2026-07-05T11:53:29.511954+00:00"},{"alias_kind":"pith_short_16","alias_value":"TEDAQYBAS7FFPWGI","created_at":"2026-07-05T11:53:29.511954+00:00"},{"alias_kind":"pith_short_8","alias_value":"TEDAQYBA","created_at":"2026-07-05T11:53:29.511954+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.31545","citing_title":"Preference-Aware Rubric Learning for Personalized Evaluation","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TEDAQYBAS7FFPWGIT5KDTXYX3E","json":"https://pith.science/pith/TEDAQYBAS7FFPWGIT5KDTXYX3E.json","graph_json":"https://pith.science/api/pith-number/TEDAQYBAS7FFPWGIT5KDTXYX3E/graph.json","events_json":"https://pith.science/api/pith-number/TEDAQYBAS7FFPWGIT5KDTXYX3E/events.json","paper":"https://pith.science/paper/TEDAQYBA"},"agent_actions":{"view_html":"https://pith.science/pith/TEDAQYBAS7FFPWGIT5KDTXYX3E","download_json":"https://pith.science/pith/TEDAQYBAS7FFPWGIT5KDTXYX3E.json","view_paper":"https://pith.science/paper/TEDAQYBA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.10028&json=true","fetch_graph":"https://pith.science/api/pith-number/TEDAQYBAS7FFPWGIT5KDTXYX3E/graph.json","fetch_events":"https://pith.science/api/pith-number/TEDAQYBAS7FFPWGIT5KDTXYX3E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TEDAQYBAS7FFPWGIT5KDTXYX3E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TEDAQYBAS7FFPWGIT5KDTXYX3E/action/storage_attestation","attest_author":"https://pith.science/pith/TEDAQYBAS7FFPWGIT5KDTXYX3E/action/author_attestation","sign_citation":"https://pith.science/pith/TEDAQYBAS7FFPWGIT5KDTXYX3E/action/citation_signature","submit_replication":"https://pith.science/pith/TEDAQYBAS7FFPWGIT5KDTXYX3E/action/replication_record"}},"created_at":"2026-07-05T11:53:29.511954+00:00","updated_at":"2026-07-05T11:53:29.511954+00:00"}