Event-Centric Memory with Query-Aware Graph Augmentation for Long-Term Conversational Agents
To address these limitations, we propose QGMem, a novel memory construction and activation framework motivated by human memory, in which experience is organized into events and query-relevant events are modeled by graph as working memory.
ProofPaper ↗
Key points
- For persistent and personalized conversational agents, memory systems can enable them to remember, update, and reason over long histories by storing past interactions and retrieving relevant information.
- QGMem converts long dialogue histories into event-indexed atomic memory units that preserve individual experiences and consolidates related units into dynamic memory traces that retain state trajectories and current states.
- To expose relational dependencies in the working memory and support conflict-aware reasoning, QGMem organizes the working memory as a local graph, which is then encoded as a graph token and provided to the LLM together with the textual working memory to improve evidence utilization during answer generation.
- Experiments across six benchmarks validate the framework and show consistent gains in retrieval, multi-hop evidence composition, conflict resolution, and ultra-long dialogue reasoning with compact contexts and moderate inference cost.
Sources (1)
- [1]Event-Centric Memory with Query-Aware Graph Augmentation for Long-Term Conversational AgentsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 01:17 PM
To address these limitations, we propose QGMem, a novel memory construction and activation framework motivated by human memory, in which experience is organized into events and query-relevant events are modeled by graph as working memory.
For persistent and personalized conversational agents, memory systems can enable them to remember, update, and reason over long histories by storing past interactions and retrieving relevant information.
Extractive summary: sentences quoted from the sources.