RAGulator: Lightweight Out-of-Context Detectors for Grounded Text Generation

TL;DR

This paper introduces RAGulator, a lightweight model based on DeBERTa, designed to efficiently detect out-of-context LLM outputs in real-time, facilitating safer enterprise deployment of RAG applications.

Contribution

The work presents a resource-efficient, high-performance out-of-context detector using minimal preprocessing, emphasizing practical deployment considerations.

Findings

DeBERTa outperforms other models in this task

The model is fast and requires no additional feature engineering

Effective with minimal resource usage

Abstract

Real-time detection of out-of-context LLM outputs is crucial for enterprises looking to safely adopt RAG applications. In this work, we train lightweight models to discriminate LLM-generated text that is semantically out-of-context from retrieved text documents. We preprocess a combination of summarisation and semantic textual similarity datasets to construct training data using minimal resources. We find that DeBERTa is not only the best-performing model under this pipeline, but it is also fast and does not require additional text preprocessing or feature engineering. While emerging work demonstrates that generative LLMs can also be fine-tuned and used in complex data pipelines to achieve state-of-the-art performance, we note that speed and resource limits are important considerations for on-premise deployment.