Large Language Models Struggle to Describe the Haystack without Human Help: Human-in-the-loop Evaluation of Topic Models
This addresses the practical challenge of using LLMs for real-world document analysis, showing they are incremental improvements over traditional models but still require human help for domain-specific tasks.
This study evaluated how effectively large language models (LLMs) help users understand large document collections compared to traditional topic models, finding that while LLMs produce more readable topics and higher win probabilities for exploration, they generate overly generic topics for domain-specific data that limit learning, and human supervision improves results but increases effort.
A common use of NLP is to facilitate the understanding of large document collections, with a shift from using traditional topic models to Large Language Models. Yet the effectiveness of using LLM for large corpus understanding in real-world applications remains under-explored. This study measures the knowledge users acquire with unsupervised, supervised LLM-based exploratory approaches or traditional topic models on two datasets. While LLM-based methods generate more human-readable topics and show higher average win probabilities than traditional models for data exploration, they produce overly generic topics for domain-specific datasets that do not easily allow users to learn much about the documents. Adding human supervision to the LLM generation process improves data exploration by mitigating hallucination and over-genericity but requires greater human effort. In contrast, traditional. models like Latent Dirichlet Allocation (LDA) remain effective for exploration but are less user-friendly. We show that LLMs struggle to describe the haystack of large corpora without human help, particularly domain-specific data, and face scaling and hallucination limitations due to context length constraints.