PhantomWiki: On-Demand Datasets for Reasoning and Retrieval Evaluation
This provides a scalable and data leakage-resistant framework for disentangled evaluation of reasoning, retrieval, and tool-use abilities in LLMs, addressing a key issue for researchers and developers in AI.
The authors tackled the problem of data leakage and inflated performance in evaluating reasoning and retrieval capabilities of large language models by proposing PhantomWiki, a pipeline that generates unique, factually consistent document corpora with question-answer pairs on demand, which they found to be surprisingly challenging for frontier LLMs.
High-quality benchmarks are essential for evaluating reasoning and retrieval capabilities of large language models (LLMs). However, curating datasets for this purpose is not a permanent solution as they are prone to data leakage and inflated performance results. To address these challenges, we propose PhantomWiki: a pipeline to generate unique, factually consistent document corpora with diverse question-answer pairs. Unlike prior work, PhantomWiki is neither a fixed dataset, nor is it based on any existing data. Instead, a new PhantomWiki instance is generated on demand for each evaluation. We vary the question difficulty and corpus size to disentangle reasoning and retrieval capabilities respectively, and find that PhantomWiki datasets are surprisingly challenging for frontier LLMs. Thus, we contribute a scalable and data leakage-resistant framework for disentangled evaluation of reasoning, retrieval, and tool-use abilities. Our code is available at https://github.com/kilian-group/phantom-wiki.