An LLM-based proxy can reproduce human relevance judgments about topic model outputs closely enough to substitute for an average human annotator, and classical LDA remains competitive under this test.
In Findings of the Association for Computational Lin- guistics: EACL 2024, pages 1760–1773, St
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
ProxAnn: Use-Oriented Evaluations of Topic Models and Document Clustering
An LLM-based proxy can reproduce human relevance judgments about topic model outputs closely enough to substitute for an average human annotator, and classical LDA remains competitive under this test.