Research
Commonsense Knowledge with Negation: A Resource to Enhance Negation Understanding
arXiv:2604.19921v1 Announce Type: new Abstract: Negation is a common and important semantic feature in natural language, yet Large Language Models (LLMs) struggle when negation is involved in natural
arXiv:2604.19921v1 Announce Type: new Abstract: Negation is a common and important semantic feature in natural language, yet Large Language Models (LLMs) struggle when negation is involved in natural language understanding tasks. Commonsense knowledge, on the other hand, despite being a well-studied topic, lacks investigations involving negation. In this work, we show that commonsense knowledge with negation is challenging for models to understand. We present a novel approach to automatically augment existing commonsense knowledge corpora with negation, yielding two new corpora containing over 2M triples with if-then relations. In addition, pre-training LLMs on our corpora benefits negation understanding.
Related
- Understanding New-Knowledge-Induced Factual Hallucinations in LLMs: Analysis and Interpretation
- Sense and Sensitivity: Examining the Influence of Semantic Recall on Long Context Code Reasoning
- How Hypocritical Is Your LLM judge? Listener-Speaker Asymmetries in the Pragmatic Competence of Large Language Models
- Cell-Based Representation of Relational Binding in Language Models
- No-Worse Context-Aware Decoding: Preventing Neutral Regression in Context-Conditioned Generation
Source: arXiv cs.CL | 2026-04-23