SynapticRAG: Enhancing Temporal Memory Retrieval in Large Language Models through Synaptic Mechanisms
This work addresses memory retrieval issues in conversational systems, bridging cognitive science and language model development, though it appears incremental as it builds on existing retrieval methods with biological inspiration.
The paper tackles the problem of degraded accuracy in Large Language Models when handling temporally distributed conversations by proposing SynapticRAG, which combines temporal association triggers with biologically-inspired synaptic propagation mechanisms. The result shows consistent improvements of up to 14.66% points across multiple metrics on four datasets in English, Chinese, and Japanese compared to state-of-the-art memory retrieval methods.
Existing retrieval methods in Large Language Models show degradation in accuracy when handling temporally distributed conversations, primarily due to their reliance on simple similarity-based retrieval. Unlike existing memory retrieval methods that rely solely on semantic similarity, we propose SynapticRAG, which uniquely combines temporal association triggers with biologically-inspired synaptic propagation mechanisms. Our approach uses temporal association triggers and synaptic-like stimulus propagation to identify relevant dialogue histories. A dynamic leaky integrate-and-fire mechanism then selects the most contextually appropriate memories. Experiments on four datasets of English, Chinese and Japanese show that compared to state-of-the-art memory retrieval methods, SynapticRAG achieves consistent improvements across multiple metrics up to 14.66% points. This work bridges the gap between cognitive science and language model development, providing a new framework for memory management in conversational systems.