ContraDoc: Understanding Self-Contradictions in Documents with Large Language Models
This addresses the limited research on LLM capabilities for detecting self-contradictions in documents, which is important for improving document analysis and reliability in AI applications, though it is incremental as it focuses on dataset creation and benchmarking.
The paper tackles the problem of understanding self-contradictions in long documents by introducing ContraDoc, the first human-annotated dataset for this task, and finds that while GPT4 performs best and can outperform humans, it remains unreliable, especially with nuanced contradictions.
In recent times, large language models (LLMs) have shown impressive performance on various document-level tasks such as document classification, summarization, and question-answering. However, research on understanding their capabilities on the task of self-contradictions in long documents has been very limited. In this work, we introduce ContraDoc, the first human-annotated dataset to study self-contradictions in long documents across multiple domains, varying document lengths, self-contradictions types, and scope. We then analyze the current capabilities of four state-of-the-art open-source and commercially available LLMs: GPT3.5, GPT4, PaLM2, and LLaMAv2 on this dataset. While GPT4 performs the best and can outperform humans on this task, we find that it is still unreliable and struggles with self-contradictions that require more nuance and context. We release the dataset and all the code associated with the experiments (https://github.com/ddhruvkr/CONTRADOC).