Adaptive Memory Admission Control for LLM Agents

Guilin Zhang; Wei Jiang; Xiejiashan Wang; Aisha Behr; Kai Zhao; Jeffrey Friedman; Xu Chu; Amine Anoun

cs.AIMar 4, 2026

Adaptive Memory Admission Control for LLM Agents

Authors: Guilin Zhang, Wei Jiang, Xiejiashan Wang, Aisha Behr, Kai Zhao, Jeffrey Friedman, Xu Chu, Amine Anoun

Organizations: Workday AI

Abstract

LLM-based agents increasingly rely on long-term memory to support multi-session reasoning and interaction, yet current systems provide little control over what information is retained. In practice, agents either accumulate large volumes of conversational content, including hallucinated or obsolete facts, or depend on opaque, fully LLM-driven memory policies that are costly and difficult to audit. As a result, memory admission remains a poorly specified and weakly controlled component in agent architectures. To address this gap, we propose Adaptive Memory Admission Control (A-MAC), a framework that treats memory admission as a structured decision problem. A-MAC decomposes memory value into five complementary and interpretable factors: future utility, factual confidence, semantic novelty, temporal recency, and content type prior. The framework combines lightweight rule-based feature extraction with a single LLM-assisted utility assessment, and learns domain-adaptive admission policies through cross-validated optimization. This design enables transparent and efficient control over long-term memory. Experiments on the LoCoMo benchmark show that A-MAC achieves a superior precision-recall tradeoff, improving F1 to 0.583 while reducing latency by 31% compared to state-of-the-art LLM-native memory systems. Ablation results identify content type prior as the most influential factor for reliable memory admission. These findings demonstrate that explicit and interpretable admission control is a critical design principle for scalable and reliable memory in LLM-based agents.

Figures & tables

Figure 1: Overview of A-MAC. Candidate memories are extracted from conversation history and evaluated using five complementary signals capturing usefulness, reliability, redundancy, temporal relevance, and persistence. A learned linear admission policy aggregates these signals into a score $\mathcal{S}(m)$ and applies a threshold $\theta$ to determine whether to admit, update, or reject a candidate.

Method	Prec.	Recall	F1	Lat.(ms)
Random	0.278	0.278	0.278	$<$ 1
MemGPT	0.316	0.333	0.324	2765 †
MemoryBank	0.368	0.583	0.452	2843 †
Equal Weights	0.362	0.694	0.476	2916 †
A-mem	0.371	1.000	0.541	3831 †
A-MAC (Ours)	0.417	0.972	0.583	2644 †

Table 1: Performance comparison on LoCoMo test set (N=225). Latency marked with † includes LLM API calls. Best F1 in bold.

Figure 2: Precision-recall tradeoff comparison. A-MAC achieves the best balance between precision and recall, occupying the upper-right region of the curve. Dashed lines indicate F1 iso-contours.

Feature	F1 w/o	$\Delta$ F1
Full Model	0.583	–
Type Prior ( $\mathcal{T}$ )	0.476	$-$ 0.107
Novelty ( $\mathcal{N}$ )	0.555	$-$ 0.028
Utility ( $\mathcal{U}$ )	0.560	$-$ 0.023
Confidence ( $\mathcal{C}$ )	0.568	$-$ 0.015
Recency ( $\mathcal{R}$ )	0.570	$-$ 0.013

Table 2: Ablation study showing performance impact of removing each feature. Type Prior (

\mathcal{T}

) is the dominant contributor.

Threshold $\theta$	Prec.	Recall	F1	Admitted
0.30	0.360	1.000	0.529	100
0.40	0.360	1.000	0.529	100
0.50	0.376	0.972	0.543	93
0.55	0.410	0.944	0.571	83
0.60	0.395	0.833	0.536	76
0.65	0.368	0.583	0.452	57

Table 3: Threshold sensitivity analysis. F1 peaks at

\theta^{*}=0.55

. Lower thresholds maximize recall at the cost of precision; higher thresholds become overly selective.

Component	Latency (ms)	Percentage
Utility $\mathcal{U}$ (LLM)	2580	97.6%
Confidence $\mathcal{C}$ (ROUGE-L)	18	0.7%
Novelty $\mathcal{N}$ (SBERT)	32	1.2%
Recency $\mathcal{R}$ (Decay)	$<$ 1	$<$ 0.1%
Type Prior $\mathcal{T}$ (Rules)	14	0.5%
Total (A-MAC)	2644	100%

Table 4: Latency breakdown by component. LLM-based Utility dominates computation time; rule-based features execute efficiently.

Domain	Samples	F1	$\Delta$ from Mean
Personal	127	0.482	+0.072
Professional	98	0.338	$-$ 0.072
Mean	225	0.410	–
Std. Dev. ( $\sigma$ )	–	0.102	–

Table 5: Cross-domain generalization performance. A-MAC transfers effectively to both domains without retuning, though Personal conversations yield higher F1 due to explicit preference patterns.

Figure 3: Cross-domain F1 performance. Personal conversations achieve higher F1 due to explicit preference statements that align with Type Prior’s strengths. The dashed line indicates mean performance across domains.

LLM Agent Memory

Large language model agent memory lets agents retain, organize, retrieve, and update information across interactions and long-running tasks.

See all 373 papers

Memory as a Controlled Process: Learned Adaptive Memory Management for LLM Agents
Jul 15, 2026Eric Hanchen Jiang, Zhi Zhang, Yuchen Wu +11Persistent Memory for Language Models Memory Management for LLM Agents
University of California Los Angeles · University of Washington · Northwestern University
Oblivion: Self-Adaptive Agentic Memory Control through Decay-Driven Activation
Mar 31, 2026Ashish Rana, Chia-Chien Hung, Qumeng Sun +2Long-Horizon LLM Agents Agent Memory Management
NEC Laboratories Europe, Heidelberg, Germany · GWDG, Georg-August-Universität Göttingen, Göttingen, Germany
AdaMem: Learning What to Remember with Adaptive Memory Policies for Personalized Agents
Jun 19, 2026Xingyu Chen, Rui Wang, Zhaopeng Tu +1LLM Agent Memory Persistent Memory for Language Models
Shanghai Jiao Tong University, Tencent · Shanghai Jiao Tong University · Tencent

Jul 15, 2026J/K move · Enter open · S save

Memory as a Controlled Process: Learned Adaptive Memory Management for LLM Agents

Eric Hanchen Jiang, Zhi Zhang, Yuchen Wu, Levina Li+10

University of California Los Angeles · University of Washington · Northwestern University

Persistent Memory for Language Models Memory Management for LLM Agents Agent Memory Management Memory-Augmented Language Model Agents

Open paper AskPDF arXiv

Large Language Model (LLM) agents increasingly rely on external memory systems to accumulate experience across tasks. Yet nearly all existing approaches, from graph-structured memories to reflective insight stores, access memory through fixed, hand-designed heuristics. We argue that this static view of memory is a core bottleneck for agentic learning because optimal memory behavior is fundamentally context-dependent. The early stages of the tasks, benefit from minimal retrieval because memory is sparse; recurring goal types benefit from plan reuse rather than generic nearest-neighbor lookup; stuck agents benefit from re-retrieval with alternative queries; and across long task streams, the memory store itself must be consolidated and pruned to remain useful. We present Memory as a Controlled Process (MemCon), a framework that models memory operations as a Markov Decision Process and learns an online policy that adaptively decides when, what, and how much to retrieve, when to inject a distilled plan, and when to consolidate or forget. MemCon is backend-agnostic: it wraps any existing memory implementation, learns from task-by-task binary feedback with no pretraining and no additional LLM calls, and uses a lightweight tabular contextual bandit with UCB exploration that converges within tens of tasks. Across 6 benchmarks, 3 agent frameworks, and 3 LLM backbones, MemCon consistently outperforms multiple memory baselines by up to 15.2 points in task success while reducing token consumption by 5--20%.

Abstract

LLM Agent Memory

Explore similar work

Memory as a Controlled Process: Learned Adaptive Memory Management for LLM Agents

Oblivion: Self-Adaptive Agentic Memory Control through Decay-Driven Activation

AdaMem: Learning What to Remember with Adaptive Memory Policies for Personalized Agents