DYNA : Dynamic Episodic Memory Networks for Augmenting Large Language Models with Temporal Knowledge Graphs in Continuous Learning
Year: 2026
Type: preprint
Abstract: Large Language Models (LLMs) struggle to incorporate new knowledge without forgetting or costly retraining. We propose DYNA, a lightweight framework that augments a frozen LLM with a temporal knowledge graph where events are nodes and temporal relations are directed, timestamped edges. The graph serves as an external, updatable memory. At query time, DYNA retrieves relevant nodes via random walks and centrality measures, then augments the LLM's response. Evaluated on three temporal recall tasks, DYNA reduces catastrophic forgetting by ~7% compared to fine-tuning and improves temporal ordering by ~5% over standard RAG. Higher graph clustering coefficients correlate with better retrieval, showing that graph structure matters. Contributions: (1) episodic memory as temporal KG, (2) retraining-free LLM augmentation, (3) graph properties as predictors of retrieval performance.
Source: arXiv (Cornell University)
Authors: Sarabadani, Ali, Tajvidiyan, Mahtab
Institution: University of Qom
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7164922151
Cites:
Cited by:
Related to: