QA-Noun: Representing Nominal Semantics via Natural Language Question-Answer Pairs

2025-11-16

AI Summary

Existing semantic analysis methods excel at modeling predicate-argument structures but inadequately capture noun-centric semantics—such as implicit roles and context-dependent relations. To address this gap, we propose QA-Noun, the first framework to systematically extend question-answering (QA) mechanisms to noun semantic modeling. It introduces nine carefully designed question templates covering both explicit syntactic roles and implicit contextual roles of nouns, enabling fine-grained sentence meaning decomposition. Our approach integrates QA-based semantic parsing, a refined annotation guideline, supervised learning models, and joint inference with QA-SRL. Experiments show that QA-Noun achieves near-perfect coverage (≈100%) of AMR noun arguments and significantly improves implicit relation identification. When jointly deployed with QA-SRL, it enhances semantic decomposition granularity by over 130% compared to baselines like FactScore, thereby bridging a critical gap in predicate-centric semantic frameworks.

Technology Category

• Natural Language Processing: Question Answering
• Knowledge Representation and Reasoning: Argumentation
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Decomposing sentences into fine-grained meaning units is increasingly used to model semantic alignment. While QA-based semantic approaches have shown effectiveness for representing predicate-argument relations, they have so far left noun-centered semantics largely unaddressed. We introduce QA-Noun, a QA-based framework for capturing noun-centered semantic relations. QA-Noun defines nine question templates that cover both explicit syntactical and implicit contextual roles for nouns, producing interpretable QA pairs that complement verbal QA-SRL. We release detailed guidelines, a dataset of over 2,000 annotated noun mentions, and a trained model integrated with QA-SRL to yield a unified decomposition of sentence meaning into individual, highly fine-grained, facts. Evaluation shows that QA-Noun achieves near-complete coverage of AMR's noun arguments while surfacing additional contextually implied relations, and that combining QA-Noun with QA-SRL yields over 130% higher granularity than recent fact-based decomposition methods such as FactScore and DecompScore. QA-Noun thus complements the broader QA-based semantic framework, forming a comprehensive and scalable approach to fine-grained semantic decomposition for cross-text alignment.

Problem

• Capturing noun-centered semantic relations via QA pairs
• Addressing limitations in current QA-based semantic frameworks
• Enhancing granularity of semantic decomposition for cross-text alignment

Innovation

• QA-Noun framework captures noun-centered semantic relations
• Nine question templates cover explicit and implicit noun roles
• Combines with QA-SRL for unified fine-grained semantic decomposition