LOKA: Conflict-Aware LLM Knowledge Update with Adaptive Knowledge Memory

Publication Date: 7/7/2026

Event: The 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026)

Reference: pp. 16689–16715

Authors: Binchi Zhang, University of Virginia; Zhengzhang Chen, NEC Laboratories America, Inc.; Zaiyi Zheng, University of Virginia; Jundong Li, University of Virginia; Haifeng Chen, NEC Laboratories America, Inc.

Abstract: Large Language Models (LLMs) have achieved remarkable success in natural language processing by encoding extensive knowledge, but their utility relies on timely updates as human knowledge keeps evolving. In this paper, we investigate the problem of LLM knowledge updates, which requires simultaneously unlearning un-wanted information and learning new knowledge. Existing approaches that tackle unlearning and learning separately encounter task conflicts and knowledge management issues when applied to comprehensive knowledge updates. In this paper, we validate our findings with theoretical analysis and empirical evidence, and propose LOKA, a conflict-aware framework for Large language mOdel Knowledge updAtes. During training, LOKA introduces an adaptive knowledge memory approach in which updated knowledge is allocated across multiple memory units. During inference, LOKA retrieves the most relevant memory unit from the knowledge memory and integrates it with the original LLM to apply updated knowledge, while a learning-based router controls the activation of the knowledge memory to improve knowledge utilization. Extensive experiments demonstrate the efficacy of LOKA in achieving accurate, flexible, and conflict-aware knowledge updates.

Publication Link: https://aclanthology.org/2026.acl-long.760/