Long-term memory enables large language model (LLM) agents to provide continuous personalized assistance by retaining user information across interactions. Existing systems support this process through fact extraction, organization, updating, and retrieval, but largely maintain memories as independent facts. This fact-centric paradigm may overlook how new facts update, constrain, or conflict with earlier ones, while leaving their applicability, role, and priority for the current request implicit. In this paper, we introduce EvoMemState, a state-centric framework that augments persistent facts with two complementary states: Relational States that capture how facts remain valid and relate to one another as memory evolves, and a Reasoning State that determines how these facts should be applied to the current request. Specifically, EvoMemState consists of three key components: (i) Ontology-Guided Memory State Induction extracts Persistent Facts and constructs Relational States that maintain their validity, scope, and evolving relations. (ii) RL-Based State Evolution internalizes memory-use experience into a policy through reinforcement learning to generate a request-specific Reasoning State. (iii) State-Aware Reasoning then combines the relevant facts with their Relational States and the generated Reasoning State to guide the personal agent. Extensive experiments on long-term dialogue and memory-use benchmarks demonstrate consistent improvements over existing memory systems in retrieval completeness and memory-grounded response accuracy. Our code and implementation details are available at https://github.com/XMUDeepLIT/EvoMemState.git.
Figures & tables
Figure 1: Memory-centric and state-centric paradigms. Our retrieval preserves updates and conflicts, while generated reasoning states specify memory applicability, role, and priority for the current request.
Figure 2: Retrieval completeness and conditional answer accuracy on the Valid Memory Selection subset of MemSyco-Bench across evaluated systems.
Figure 3: Effect of fixed self-reflection on answer accuracy when complete memory evidence is available.
Figure 4: Overview of EvoMemState for personal agents. Ontology-guided induction transforms accumulated personal information into Persistent Facts and Relational States. RL-based state evolution learns a Memory State Evolution Policy from memory-use experience and generates a request-specific Reasoning State. State-Aware Reasoning combines the Relational States and Reasoning State to guide the personal agent’s response.
Memory System
MemSyco-Bench
MemTrapBench
LoCoMo- Refined
LoCoMo
Overall Avg. ↑
WTU Acc. ↑
WTU Syco. ↓
HTU Acc. ↑
HTU Use ↑
Avg. ↑
Safety Score ↑
CogBias Acc. ↑
Bound. Acc. ↑
Trauma Acc. ↑
Avg. ↑
Judge Score ↑
Token F1 ↑
Judge Score ↑
Token F1 ↑
Qwen3-8B
NaiveRAG ( Lewis et al., 2020 )
36.3
55.3
51.6
64.0
42.4
60.2
63.9
77.0
93.4
73.6
39.0
38.4
34.4
35.8
47.4
MemGPT ( Packer et al., 2023 )
29.0
66.7
54.0
67.0
39.0
62.2
66.0
64.5
52.4
61.3
45.7
40.0
36.8
37.0
45.7
MemoryBank ( Zhong et al., 2024 )
37.5
57.6
54.4
67.1
44.2
64.9
63.7
72.1
82.3
70.7
44.9
44.0
39.9
41.0
49.9
Mem0 ( Chhikara et al., 2025 )
25.9
47.8
52.0
63.8
36.3
56.8
69.3
81.9
85.5
73.4
25.7
26.9
21.4
23.8
39.2
Table 1: Main results on four benchmarks with Qwen3-8B and GPT-4o-mini. ↑ and ↓ denote metrics to maximize and minimize, respectively. Bold and underlined values denote the best and second best results for each backbone. WTU/HTU: When/How to Use memory; Syco.: sycophancy rate; Use: correct use.
Figure 5: Complete retrieval rate and accuracy on questions with complete evidence using Qwen3-8B. Average complete retrieval rate reports the percentage of questions with all required evidence retrieved. Acc@complete reports accuracy on LoCoMo-Refined and MemSyco-Bench for questions with all required evidence retrieved.
Method
MemSyco Avg.
LoCoMo-R LLM-J.
LoCoMo LLM-J.
MemTrap Avg.
Overall
EvoMemState (Ours)
57.6
57.4
45.5
79.1
59.9
w/o Relational State
48.7
54.9
45.2
72.7
55.4
w/o Reasoning State
49.6
55.8
42.5
64.1
53.0
w/o RL-Based State Evolution
49.7
55.4
45.0
62.7
53.2
Table 2: Ablation results for EvoMemState with Qwen3-8B. The variants remove Relational State, Reasoning State or RL-Based State Evolution.
Figure 6: Cross-system transfer of the Memory State Evolution Policy, averaged over MemSyco-Bench, MemTrapBench, and LoCoMo-Refined. Base and +Reasoning State denote results without and with generated reasoning states; Δ is the gain.
Appendix figures & tables14 assets
Supplementary material from the paper’s appendix.
Appendix
Figure 7: Ontology-Guided Memory State Induction. Each dialogue session yields Persistent Facts, concept edits, and memory relations. CREATE adds a concept, MERGE combines duplicate concepts, and SPLIT introduces more specific concepts. Persistent Facts remain traceable while their Relational States evolve with the ontology.
Method
MemSyco-Bench
LoCoMo-Refined
MemTrapBench
Scope Acc. ↑
Objective Acc. ↑
Personalized Acc. ↑
Preference Change Acc. ↑
Preference Conflict Acc. ↑
LLM-Judge Score ↑
Token F1 ↑
Safety Acc. ↑
Cognitive Bias Acc. ↑
Task Boundary Acc. ↑
Trauma Acc. ↑
Avg. ↑
Mem0
14.3
34.7
57.7
54.5
22.3
25.7
26.9
56.8
69.3
81.9
85.5
73.4
+ Self-Recheck ( Hu et al., 2026 )
13.0
43.3
60.5
64.7
25.0
23.4
12.6
75.1
55.5
76.4
94.6
75.4
+ MemGate ( Zhang et al., 2026a )
13.7
39.0
51.0
51.2
23.7
24.2
11.4
54.6
40.9
78.7
93.5
66.9
+ Dynamic Partition ( Hannoon et al., 2026 )
11.7
39.0
56.5
53.8
8.7
20.8
10.6
69.2
44.6
78.4
93.1
71.3
+ Reasoning State (Ours)
11.0
39.3
48.8
72.7
42.0
23.9
24.9
61.1
72.3
81.9
93.8
77.3
Appendix
Table 3: Comparison of memory use strategies with Mem0 across three benchmarks. Dynamic Rules (Ours) uses the learned Memory State Evolution Policy. Bold and underlined values indicate the best and second best results, respectively.
Figure 8: Case analysis of the two limitations identified in our preliminary studies. The upper panel shows that retrieving an individually relevant fact may omit related updates and constraints required to reconstruct the current memory state. The lower panel shows that even complete evidence may be used incorrectly when its applicability, role, and priority are not determined for the current request.
Figure 9: Performance with response model guidance (Self-Guide) and learned memory rules (Reasoning State) under Qwen3-8B and GPT-4o-mini. LoCoMo and LoCoMo-Refined report LLM judge scores. Differences from Self-Guide are reported in points on the percentage scale.
Figure 10: Examples of Reasoning States expressed as dynamic memory rules for preference updates, evidence priority, and scope control. Each case presents the relevant evidence, generated rules, and response. Text is condensed from recorded outputs.
Figure 11: Retrieval hit rate and average retrieved context length on LoCoMo-Refined. Each point represents a memory system, and retrieval cost is measured in tokens.
Method
Retrieval Hit Rate
Average Coverage
Complete Retrieval Rate
EvoMemState (Ours)
0.816
0.747
0.680
MemoryBank
0.789
0.727
0.669
A-MEM
0.789
0.724
0.662
NaiveRAG
0.684
0.615
0.564
MemGPT
0.453
0.384
0.323
LightMem
0.166
0.144
0.129
Appendix
Table 4: Gold evidence recovery on LoCoMo-Refined. Hit rate counts questions with at least one gold item retrieved, coverage averages the fraction of gold items retrieved per question, and complete retrieval rate counts questions with every gold item retrieved. All three metrics are averaged across questions and reported on a 0–1 scale.
Figure 12: Statistics of memory stores on MemSyco-Bench. The panels report average entry counts, edge counts, and entries per thousand characters. Flat stores use entry counts as an edge count proxy. † includes cross note links in A-MEM. ‡ denotes density estimated from retrieved entries and context length.
Figure 13: Complete retrieval rate and Acc@complete with and without memory guidance. Acc@complete is evaluated on questions with all required evidence retrieved. Arrows report changes in percentage points.
Table 5: Prompt for extracting memories and updating the ontology.
Table 6: Prompt for generating a request-specific Reasoning State as dynamic memory rules.
Figure 20
Figure 14: Hyperparameter configurations for the memory baselines. All methods share the answer model, memory-construction LLM, embedding model, and the number of memories shown to the answer model.
Table 7: Answer-generation prompts for LoCoMo, LoCoMo-Refined, MemTrap, and MemSyco.