Memory-augmented LLM agents maintain external memory banks to support long-horizon interaction, yet most existing systems treat construction, retrieval, and utilization as isolated subroutines. This creates two coupled challenges: strategic blindness on the forward path of the memory cycle, where construction and retrieval are driven by local heuristics rather than explicit strategic reasoning, and sparse, delayed supervision on the backward path, where downstream failures rarely translate into direct repairs of the memory bank. To address these challenges, we propose MemMA, a plug-and-play multi-agent framework that coordinates the memory cycle along both the forward and backward paths. On the forward path, a Meta-Thinker produces structured guidance that steers a Memory Manager during construction and directs a Query Reasoner during iterative retrieval. On the backward path, MemMA introduces in-situ self-evolving memory construction, which synthesizes probe QA pairs, verifies the current memory, and converts failures into repair actions before the memory is finalized. Extensive experiments on LoCoMo show that MemMA consistently outperforms existing baselines across multiple LLM backbones and improves three different storage backends in a plug-and-play manner. Our code is publicly available at https://github.com/ventr1c/memma.
Figures & tables
Figure 1: Two challenges in leveraging the memory cycle effect.
Method
F1
B1
ACC
Static Baseline
22.64
17.24
52.60
Unguided Active
23.49
18.36
54.60
Strategic Active
24.78
17.73
59.21
Table 1: Preliminary analysis results (%) in LoCoMo dataset, GPT-4o-mini is the backbone LLM.
Figure 2: Overview of MemMA .
Model
Method
Multi-Hop
Temporal
Open-Domain
Single-Hop
Overall
F1
B1
ACC
F1
B1
ACC
F1
B1
ACC
F1
B1
ACC
F1
B1
ACC
GPT
Full Text
29.41
21.16
43.75
29.95
19.33
51.35
18.25
19.56
61.54
41.45
29.96
74.29
34.13
24.63
61.18
Naive RAG
15.84
9.50
31.25
17.30
12.36
35.14
17.40
16.65
46.15
39.32
30.35
58.57
27.14
20.41
46.05
LangMem
12.55
9.22
25.00
15.23
11.53
21.62
14.91
14.03
38.46
23.52
17.59
35.71
18.46
14.05
30.26
A-Mem
15.56
10.88
31.25
55.01
42.40
51.35
18.18
15.27
53.85
42.72
32.43
62.86
37.90
28.85
52.63
LightMem
33.74
29.33
65.62
59.76
51.12
78.38
31.85
24.23
76.92
43.88
34.68
78.57
44.58
36.66
75.66
Table 2: Results on LoCoMo across four question categories (multi-hop, temporal, open-domain, single-hop). We report F1, B1, and ACC (%). Best results are in bold . GPT-4o-mini and Claude-Haiku-4.5 are backbones; GPT-4o-mini is the answer agent. MemMALM uses LightMem as storage backend.
Method
F1
B1
ACC
Single-Agent
22.64
17.24
52.60
MemMASA
23.64
12.94
84.87
A-Mem
37.90
28.85
52.63
MemMAAM
46.23
35.13
78.29
LightMem
44.58
36.66
75.66
MemMALM
49.40
38.28
81.58
Table 3: Flexibility across backends on LoCoMo under GPT-4o-mini. Best results per backend are in bold .
Figure 3: Ablation studies of MemMASA under GPT-4o-mini and Claude-Haiku-4.5 on LoCoMo.
Figure 4: Impact of retrieval budget k of MemMA under both GPT-4o-mini and Claude-Haiku-4.5.
Figure 5: Impact of refinement budget H of MemMA .
Method
Coverage
Hit@ ≥ 80%
NaiveRAG
42.2
24.0
A-Mem
51.0
26.0
LightMem
61.9
39.0
MemMA LM , E0
65.2
45.2
MemMA LM , EH
66.4
45.9
Table 4: Retrieval-process diagnostics (%) on LoCoMo. E0 and EH denote the initial and final accumulated evidence, respectively.
Method
ACC
FullContext
50.0
NaiveRAG
58.0
A-Mem
72.0
LightMem
82.0
MemMA LM
94.0
Table 5: Results on the 20% subset of LongMemEval-S.
Method
Memory Scope
Pass Rate
Vanilla
None
62.0
MemMA /E
Within-task
66.0
MemMA
Within-task
72.0
MemMA /E
Within + cross-task
72.0
MemMA
Within + cross-task
76.0
Table 6: Results on SWE-bench-Verified-mini.
Appendix figures & tables12 assets
Supplementary material from the paper’s appendix.
Appendix
Table 7: The prompt template used for LLM-as-a-Judge evaluation.
Type
Example Question
Example Answer
Single-hop
What type of support group did I tell Melanie I attended recently?
An LGBTQ support group
Multi-hop
What is Melanie’s hobby for creative expression and relaxation, and when did she create the specific piece she showed me?
Melanie paints as her hobby for creative expression and relaxation. She painted a lake sunrise last year that she showed me.
Temporal
On what date and time did I have the conversation with Melanie about attending the LGBTQ support group and my career interests in counseling?
At 1:56 pm on May 8, 2023
Appendix
Table 8: Synthetic QA probe types used during probe generation in LoCoMo, with representative examples from the generated probe data.
Probe Model
F1
B1
ACC
Claude-Haiku-4.5
44.98
35.69
74.34
Claude-Sonnet-4.5
43.30
32.74
74.34
Claude-Opus-4.5
45.10
35.66
76.97
Appendix
Table 9: Impact of probe generation model on MemMALM with Claude-Haiku-4.5 as the construction backbone. Best results are in bold .
Probe Model
Avg. Q Len.
Avg. A Len.
One-word / ≤ 3
Single-hop
Multi-hop
Temporal
Haiku
18.48
19.44
4 / 15
55
25
15
Sonnet
15.42
11.13
11 / 33
64
16
15
Opus
17.38
21.55
4 / 8
58
26
11
Appendix
Table 10: Probe statistics across generation models. “One-word / ≤ 3” counts one-word and short ( ≤3 words) answers out of 95 total per model. Question type counts follow the taxonomy in Appendix E.1 .
Model
Question
Answer
Haiku
What has the support group I attended done for my personal development and self-acceptance?
The support group has made me feel accepted and given me courage to embrace myself.
Sonnet
What did the LGBTQ support group help me feel that gave me courage to embrace myself?
Accepted
Opus
How has attending the LGBTQ support group influenced my personal growth and willingness to be open about my identity?
The support group has been a safe space that made me feel accepted, giving me the courage to embrace myself and be more open about my identity in other areas of life.
Appendix
Table 11: Representative probe QA pairs from the same dialogue session. Sonnet’s single-word answer tests only keyword presence, while Haiku and Opus require multi-attribute recall.
Table 12: The prompt template used for Meta-Thinker construction guidance.
Table 13: The prompt template used for Meta-Thinker answerability checking (Part 1).
Table 14: The prompt template used for Meta-Thinker answerability checking (Part 2).
Table 15: The prompt template used for Query Reasoner πr to generate orthogonal query uh+1 .
Table 16: The prompt template used for evidence-grounded repair in self-evolution (Part 1).
Table 17: The prompt template used for evidence-grounded repair in self-evolution (Part 2: output format and examples).
Table 18: The prompt template used for semantic consolidation (deduplication) in self-evolution.
Memory is essential for enabling LLM-based agents to maintain coherent, personalized behavior over long-horizon interactions. However, existing memory systems share a fundamental limitation: they never proactively test their own memory, repairing it only after real queries expose weaknesses. This reactive paradigm means every retrieval failure corresponds to a real interaction in which the cost has already been paid. We propose MemDream, a framework that enables self-probing memory evolution for LLM agents. Our framework periodically enters offline dream cycles where three specialized agents (Dreamer, Analyst, Consolidator) collaboratively probe, diagnose, and repair the memory graph before failures occur. A policy trained via Group Relative Policy Optimization learns which repair operations produce durable retrieval improvements, while a soft decay mechanism provides reversible forgetting driven by the same anticipatory signal. Experiments on LoCoMo and MemoryAgentBench demonstrate that MemDream improves answer F1 by 4.5 points on LoCoMo and achieves a 9.1-point higher overall score on MAB over the strongest reactive-evolution baselines.
Mingfei Lu, Mengjia Wu, Runsong Jia +2
Australian Artificial Intelligence Institute (AAII) University of Technology Sydney
Despite recent progress, LLM agents still struggle with reasoning over long interaction histories. While current memory-augmented agents rely on a static retrieve-then-reason paradigm, this rigid pipeline design prevents them from dynamically adapting memory access to intermediate evidence discovered during inference. To bridge this gap, we propose MRAgent, a framework that combines an associative memory graph with an active reconstruction mechanism. We represent memory as a Cue-Tag-Content graph, where associative tags serve as semantic bridges connecting fine-grained cues to memory contents. Operating on this structure, our active reconstruction mechanism integrates LLM reasoning directly into memory access, allowing the agent to iteratively explore and prune retrieval paths based on accumulated evidence. This ensures that memory retrieval is dynamically adapted to the reasoning context while avoiding combinatorial explosion caused by unconstrained expansion. Experiments on the LoCoMo benchmark and LongMemEval benchmark demonstrate significant improvements over strong baselines (up to 23%), while substantially reducing token and runtime cost, highlighting the effectiveness of active and associative reconstruction for long-horizon memory reasoning.
Large Language Models (LLMs) show promise as tool-using agents but remain limited in long-horizon tasks that require remembering, organizing, and reusing knowledge. Prior memory approaches aim to resolve the situation, but mainly focus on storing factual information. Recent work on procedural memory improves task reuse, yet often reduces to replaying past successes without addressing failure cases or online scalability. We introduce a unified and automatic memory framework that integrates semantic, episodic, and procedural memory in a bi-level design combining short-term and long-term stores. A multi-agent architecture with actor, memory, and critic agents enables automatic memory generation, reward annotation, and adaptive retrieval. Long-term memory is managed through reward-based evaluation, merging, and pruning, ensuring scalability and continual improvement. Experiments across various environments show that our approach improves robustness and success on long multi-turn tasks compared to existing baselines. This work highlights the importance of comprehensive, adaptive memory for advancing LLM-based agents.