RelationalFactQA: A Benchmark for Evaluating Tabular Fact Retrieval from Large Language Models

TL;DR

This paper introduces RelationalFactQA, a benchmark for evaluating large language models' ability to retrieve and generate structured, multi-record tabular factual data, revealing significant current limitations.

Contribution

We present a new benchmark, RelationalFactQA, designed to evaluate the structured fact retrieval capabilities of LLMs, highlighting their struggles with complex, multi-record outputs.

Findings

State-of-the-art LLMs achieve less than 25% accuracy on relational outputs.

Performance declines as output size and complexity increase.

Current models exhibit significant failure modes in structured factual generation.

Abstract

Factuality in Large Language Models (LLMs) is a persistent challenge. Current benchmarks often assess short factual answers, overlooking the critical ability to generate structured, multi-record tabular outputs from parametric knowledge. We demonstrate that this relational fact retrieval is substantially more difficult than isolated point-wise queries, even when individual facts are known to the model, exposing distinct failure modes sensitive to output dimensionality (e.g., number of attributes or records). To systematically evaluate this under-explored capability, we introduce RelationalFactQA, a new benchmark featuring diverse natural language questions (paired with SQL) and gold-standard tabular answers, specifically designed to assess knowledge retrieval in a structured format. RelationalFactQA enables analysis across varying query complexities, output sizes, and data characteristics. Our experiments reveal that even state-of-the-art LLMs struggle significantly, not exceeding 25% factual accuracy in generating relational outputs, with performance notably degrading as output dimensionality increases. These findings underscore critical limitations in current LLMs' ability to synthesize structured factual knowledge and establish RelationalFactQA as a crucial resource for measuring future progress in LLM factuality.