{"paper":{"title":"Bridging the Semantic Gap for Categorical Data Clustering via Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"LLM-generated semantic descriptions of categorical values, when fused into embeddings, measurably improve clustering quality over standard methods.","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Xin Liao, Yiqun Zhang, Yiu-ming Cheung, Zihua Yang","submitted_at":"2026-01-03T11:37:46Z","abstract_excerpt":"Qualitative data are widespread in domains such as healthcare, marketing, and bioinformatics, where clustering offers a fundamental tool for pattern discovery. A core difficulty of qualitative-data clustering lies in measuring similarity among attribute values that carry no inherent ordering or distance. To recover such relationships, existing studies typically rely on within-dataset co-occurrence statistics. This statistical route, however, becomes unreliable once the sample size is small, and the semantic context of each value is therefore left underexploited. Motivated by this limitation, t"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Experiments on eight benchmark datasets demonstrate consistent improvements over seven representative counterparts, with gains of 19-27%.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That LLM-generated descriptions of attribute values provide reliable, unbiased semantic knowledge that meaningfully complements the original categorical metric space without introducing hallucinations or domain mismatches.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"ARISE integrates LLM-generated semantic embeddings with categorical data representations to bridge the similarity gap, achieving 19-27% gains over seven baselines on eight benchmark datasets.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"LLM-generated semantic descriptions of categorical values, when fused into embeddings, measurably improve clustering quality over standard methods.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"ec958ae73bfe1390e2d0422f540ab0139e49963b22e5b176e29ea22b6424bd37"},"source":{"id":"2601.01162","kind":"arxiv","version":3},"verdict":{"id":"52a8420f-9de6-4304-b091-904471b7c303","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T18:02:03.643604Z","strongest_claim":"Experiments on eight benchmark datasets demonstrate consistent improvements over seven representative counterparts, with gains of 19-27%.","one_line_summary":"ARISE integrates LLM-generated semantic embeddings with categorical data representations to bridge the similarity gap, achieving 19-27% gains over seven baselines on eight benchmark datasets.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That LLM-generated descriptions of attribute values provide reliable, unbiased semantic knowledge that meaningfully complements the original categorical metric space without introducing hallucinations or domain mismatches.","pith_extraction_headline":"LLM-generated semantic descriptions of categorical values, when fused into embeddings, measurably improve clustering quality over standard methods."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2601.01162/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"}