Paraphrasing in Affirmative Terms Improves Negation Understanding

2024/06/11 by MohammadHossein Rezaei, Rezaei, MohammadHossein, Eduardo Blanco +1 · 1 citation
Neuroscience · Psychology · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Language, Metaphor, and Cognition #Neurobiology of Language and Bilingualism #Second Language Acquisition and Learning

paper · pdf · doi:10.48550/arxiv.2406.07492

openalex publication_date 2024/06/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Negation is a common linguistic phenomenon. Yet language models face challenges with negation in many natural language understanding tasks such as question answering and natural language inference. In this paper, we experiment with seamless strategies that incorporate affirmative interpretations (i.e., paraphrases without negation) to make models more robust against negation. Crucially, our affirmative interpretations are obtained automatically. We show improvements with CondaQA, a large corpus requiring reasoning with negation, and five natural language understanding tasks.

Cited by

• Thunder-NUBench: A Benchmark for LLMs' Sentence-Level Negation Understanding

Related

• Leveraging Affirmative Interpretations from Negation Improves Natural Language Understanding
• A Question-Answer Driven Approach to Reveal Affirmative Interpretations from Verbal Negations
• Making Language Models Robust Against Negation
• CONDAQA: A Contrastive Reading Comprehension Dataset for Reasoning about Negation
• This is not a Dataset: A Large Negation Benchmark to Challenge Large Language Models