A prompt-based LLM pipeline infers entity type hierarchies, conceptual attributes, and inter-type relationships from raw tables, and reports improved or comparable quality over embedding baselines on two tabular benchmarks.
Dataset discovery and exploration: A survey,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DB 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Conceptual Schema Inference for Tabular Datasets using Large Language Models
A prompt-based LLM pipeline infers entity type hierarchies, conceptual attributes, and inter-type relationships from raw tables, and reports improved or comparable quality over embedding baselines on two tabular benchmarks.