Memory-Augmented Knowledge Fusion with Safety-Aware Decoding for Domain-Adaptive Question Answering
Lei Fu, Xiang Chen, Kaige Gao Xinyue Huang, Kejian Tong
TL;DR
The paper addresses the challenge of building trustworthy domain-adaptive QA systems by integrating heterogeneous knowledge sources while enforcing safety. It introduces KARMA, a modular framework with Multi-Source Knowledge Fusion, a Gated Memory Unit, and a Safety-Aware Controllable Decoder, trained via a multi-objective loss that balances fluency, safety, and knowledge alignment. Empirical results on a proprietary elderly-service QA dataset show substantial gains over strong baselines in accuracy, safety, and knowledge relevance, demonstrating the practicality of the approach. These contributions enable more reliable and context-aware QA in sensitive service domains, with potential for real-world deployment in elder care and government-program guidance.
Abstract
Domain-specific question answering (QA) systems for services face unique challenges in integrating heterogeneous knowledge sources while ensuring both accuracy and safety. Existing large language models often struggle with factual consistency and context alignment in sensitive domains such as healthcare policies and government welfare. In this work, we introduce Knowledge-Aware Reasoning and Memory-Augmented Adaptation (KARMA), a novel framework designed to enhance QA performance in care scenarios. KARMA incorporates a dual-encoder architecture to fuse structured and unstructured knowledge sources, a gated memory unit to dynamically regulate external knowledge integration, and a safety-aware controllable decoder that mitigates unsafe outputs using safety classification and guided generation techniques. Extensive experiments on a proprietary QA dataset demonstrate that KARMA outperforms strong baselines in both answer quality and safety. This study offers a comprehensive solution for building trustworthy and adaptive QA systems in service contexts.