CLOct 18, 2024

Critical Questions Generation: Motivation and Challenges

arXiv:2410.14335v125 citationsh-index: 5CoNLL
Originality Incremental advance
AI Analysis

This work addresses misinformation mitigation for users of argumentative texts, but it is incremental as it builds on existing argumentation theory and LLM capabilities.

The paper tackles the problem of misinformation by proposing Critical Questions Generation, a task to generate questions that reveal blind spots in arguments, and finds that while LLMs can generate reasonable critical questions, they have significant room for improvement.

The development of Large Language Models (LLMs) has brought impressive performances on mitigation strategies against misinformation, such as counterargument generation. However, LLMs are still seriously hindered by outdated knowledge and by their tendency to generate hallucinated content. In order to circumvent these issues, we propose a new task, namely, Critical Questions Generation, consisting of processing an argumentative text to generate the critical questions (CQs) raised by it. In argumentation theory CQs are tools designed to lay bare the blind spots of an argument by pointing at the information it could be missing. Thus, instead of trying to deploy LLMs to produce knowledgeable and relevant counterarguments, we use them to question arguments, without requiring any external knowledge. Research on CQs Generation using LLMs requires a reference dataset for large scale experimentation. Thus, in this work we investigate two complementary methods to create such a resource: (i) instantiating CQs templates as defined by Walton's argumentation theory and (ii), using LLMs as CQs generators. By doing so, we contribute with a procedure to establish what is a valid CQ and conclude that, while LLMs are reasonable CQ generators, they still have a wide margin for improvement in this task.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes