When to Forget: A Memory Governance Primitive
Summary: arXiv:2604.12007v1 Announce Type: new
Abstract: Agent memory systems accumulate experience but currently lack a principled operational metric for memory quality governance — deciding which memories to trust, suppress, or deprecate as the agent’s task distribution shifts.
Write-time importance scores are static; dynamic management systems use LLM judgment or structural heuristics rather than outcome feedback. This paper proposes Memory Worth (MW): a two-counter per-memory signal that tracks how often a memory co-occurs with successful versus failed outcomes, providing a lightweight, theoretically grounded foundation for staleness detection, retrieval suppression, and deprecation decisions.
We prove that MW converges almost surely to the conditional success probability p+(m) = Pr[y_t = +1 | m in M_t] — the probability of task success given that memory m is retrieved — under a stationary retrieval regime with a minimum exploration condition. Importantly, p+(m) is an associational quantity, not a causal one: it measures outcome co-occurrence rather than causal contribution.
We argue this is still a useful operational signal for memory governance, and we validate it empirically in a controlled synthetic environment where ground-truth utility is known:
- After 10,000 episodes, the Spearman rank-correlation between Memory Worth and true utilities reaches rho = 0.89 +/- 0.02 across 20 independent seeds.
- In contrast, systems that never update their assessments yield rho = 0.00.
A retrieval-realistic micro-experiment with real text and neural embedding retrieval (all-MiniLM-L6-v2) further shows stale memories crossing the low-value threshold (MW = 0.17) while specialist memories remain high-value (MW = 0.77) across 3,000 episodes. The estimator requires only two scalar counters per memory unit and can be added to architectures that already log retrievals and episode outcomes.
Conclusion
The introduction of Memory Worth (MW) marks a significant advancement in memory governance for agent systems. By utilizing a straightforward two-counter framework, agents can dynamically assess the value of their memories in relation to task success, addressing a crucial gap in existing methodologies. This development not only enhances the operational efficiency of agents but also lays the groundwork for future research in adaptive memory systems.