Online Boundary-Aware Memory for Case-Based Reasoning Agents

Zheng Dong, Luming Shang

Keywords: Contrastive Explanation, Interpretable Decision Knowledge, Online Agent Memory, Decision Boundaries, Case-Based Reasoning

TL;DR: We introduce OBAM, an online boundary-aware memory architecture that achieves LLM agents continual improvement in streaming case-based reasoning by accumulating contrasting epistemological principles near decision boundaries over time.

Abstract: Large language model (LLM) agents increasingly operate in streaming case-based reasoning (CBR) settings, where continuous improvement from past experience is crucial. Existing methods achieve this by storing past cases and retrieving similar ones as few-shot examples. This strategy fails near decision boundaries, where highly similar cases have conflicting outcomes and the discriminative factors are not explicit. From an epistemological perspective, the agent possesses evidence but lacks justified knowledge of why similar cases diverge, leaving it unable to distinguish between conflicting precedents. We introduce Online Boundary-Aware Memory (OBAM), an agent memory architecture that discovers and refines decision-boundary knowledge progressively from the case stream. Unlike offline boundary analysis over a static dataset, OBAM detects boundaries online as the agent encounters contrasting cases and stores structured boundary memory entries encoding shared patterns and discriminative rules. These entries are refined as evidence accumulates, transforming ambiguous case experience into reusable decision knowledge. Across legal, medical, and fraud reasoning tasks, OBAM outperforms in-context learning and state-of-the-art agent memory baselines, demonstrating the value of boundary-aware online memory for continual improvement in streaming CBR agents.

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 82