Enhancing Frame Detection with Retrieval Augmented Generation

2025-02-17

AI Summary

This work addresses the limitations of existing frame detection approaches—namely, their reliance on explicit target span annotations and poor generalization. To this end, we propose RCIF, the first Retrieval-Augmented Generation (RAG)-based method for frame detection. RCIF performs end-to-end frame identification in three stages: frame embedding learning, candidate frame retrieval, and fine-grained ranking-based recognition—eliminating the need for token-level target annotations. Its key contributions include: (i) the first application of RAG to frame detection; (ii) the design of structured frame representations that enable cross-lexical generalization for NL2SPARQL tasks; and (iii) unified evaluation across zero-shot, few-shot, and fine-tuning settings. On FrameNet 1.5 and 1.7, RCIF achieves state-of-the-art performance, significantly reduces search space complexity, and improves robustness. This work establishes a novel paradigm for frame identification and knowledge query translation in semantic parsing.

Technology Category

• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Computer Vision: Image and Video Retrieval
• Knowledge Representation and Reasoning: Argumentation

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Recent advancements in Natural Language Processing have significantly improved the extraction of structured semantic representations from unstructured text, especially through Frame Semantic Role Labeling (FSRL). Despite this progress, the potential of Retrieval-Augmented Generation (RAG) models for frame detection remains under-explored. In this paper, we present the first RAG-based approach for frame detection called RCIF (Retrieve Candidates and Identify Frames). RCIF is also the first approach to operate without the need for explicit target span and comprises three main stages: (1) generation of frame embeddings from various representations ; (2) retrieval of candidate frames given an input text; and (3) identification of the most suitable frames. We conducted extensive experiments across multiple configurations, including zero-shot, few-shot, and fine-tuning settings. Our results show that our retrieval component significantly reduces the complexity of the task by narrowing the search space thus allowing the frame identifier to refine and complete the set of candidates. Our approach achieves state-of-the-art performance on FrameNet 1.5 and 1.7, demonstrating its robustness in scenarios where only raw text is provided. Furthermore, we leverage the structured representation obtained through this method as a proxy to enhance generalization across lexical variations in the task of translating natural language questions into SPARQL queries.

Problem

• Explores Retrieval-Augmented Generation for frame detection.
• Develops RCIF method without explicit target span.
• Enhances generalization in translating natural language to SPARQL.

Innovation

• Retrieval-Augmented Generation for frames
• Frame embeddings from representations
• State-of-the-art FrameNet performance