- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes MemAgent, which formulates LLM agent memory management as a routing problem over heterogeneous memory providers. It manages 13 long-term memory providers and makes decisions at three stages: BEGIN selects a provider for retrieval, IN decides whether to inject short-term working memory, and END selects providers for storing the current trajectory. The router is trained via reward-filtered SFT using data collected through forced exploration, with supervision signals based on task correctness, LLM-as-a-Judge scores, and downstream memory reuse. Experiments on GAIA, WebWalkerQA, and xBench-DS show that MemAgent achieves an average accuracy of 75.7%, a 10.0 percentage-point gain over the no-memory baseline (65.7%), outperforming individual memory providers and several routing and memory-aggregation baselines.
Critical strengths
- Practical problem. Different memory representations may suit different tasks, making the selection of appropriate memory systems a meaningful problem beyond designing yet another memory representation.
- Comprehensive system design. The BEGIN, IN, and END stages cover retrieval, within-task working memory, and selective storage, providing a more complete approach than methods focused solely on retrieval.
- Clear presentation. The paper is generally well written and easy to follow.
Critical weaknesses
- The proposed approach is straightforward, but its novelty is limited. It appears to mainly bring existing components together within a unified framework, which does not necessarily constitute a substantial methodological contribution. In addition, the method is not compared against recent approaches to learned memory management and adaptive memory design, such as Memory-R1, MemRL, or other relevant methods.
- Potential overlap between memory construction and evaluation. The paper states that the same 53 GAIA Level-1 tasks are used both to build the memory pool and for evaluation. The authors should clarify whether the memory available when evaluating a task contains its own trajectory, answer, or highly similar content. How do they ensure that the reported gains reflect cross-task transfer rather than direct reuse of previously seen tasks? Results excluding Level 1 should also be reported to assess performance under a stricter held-out setting.
- Insufficient statistical validation and cost accounting. The main results are reported as single accuracy numbers, without multiple runs, confidence intervals, or significance tests. Given the limited number of benchmark tasks, some gains may depend on a small number of examples. In addition, the reported overhead of approximately 1.2 seconds per task measures routing latency, not necessarily the total cost of the full memory lifecycle. Since all 13 long-term memory providers require LLM calls during storage, the authors should clarify whether the reported overhead of less than 0.3% excludes memory construction, trajectory summarization, provider updates, and training costs.
Questions for authors
See the weaknesses above.
4 comments
Response to Reviewer DQ8M (2/2)
Q3. "Potential overlap between memory construction and evaluation. ... The authors should clarify whether the memory available when evaluating a task contains its own trajectory, answer, or highly similar content. ... Results excluding Level 1 should also be reported ..."
A3. Following MemEvolve [1], Appendix A.2 separates these two settings: "we therefore treat GAIA L1 as an experience-reuse setting rather than a strict held-out generalization test," while "the remaining 112 GAIA L2/L3 tasks and all WebWalkerQA and xBench-DS tasks are disjoint from training and serve as held-out generalization benchmarks."
To check whether retrieved memories contain same-task content or cross-task know-how, we audited the BEGIN memory injected on every task using an independent judge from a different model family (Gemini-2.5-Flash):
| GAIA L1 (seen) | GAIA L2/L3 | WebWalkerQA | xBench-DS | |
|---|---|---|---|---|
| Memory contains the current question verbatim | 11–87% | 0–2.7% | 0% | 0% |
| Gold answer string appears in memory | 23–34% | 1.4–7.2% | 0–0.6% | 0–2.7% |
| Judge: procedural know-how | 64–79% | 85–100% | 98% | 98–100% |
Mostly coincidental matches of common words. Ranges are across runs. Same-task overlap occurs only on L1. On the 382 held-out tasks, retrieved memories are procedural know-how from other tasks. Results on the 382 held-out tasks confirm all conclusions (accuracy %):
| Method | GAIA L2+L3 | WebWalkerQA | xBench-DS | Avg. |
|---|---|---|---|---|
| No Memory | 51.8 | 69.4 | 69.0 | 63.4 |
| ExpeL (best single provider by avg.) | 56.2 | 77.1 | 74.0 | 69.1 |
| SkillWeaver | 59.8 | 70.0 | 77.0 | 68.9 |
| Random Routing | 53.6 | 72.4 | 71.0 | 65.6 |
| MemAgent w/o training | 59.8 | 70.6 | 70.0 | 66.8 |
| MemAgent | 63.4 | 79.4 | 78.0 | 73.6 |
MemAgent raises the held-out average from 63.4% to 73.6% (+10.2 pp) and ranks first on every split. Table 1 already shows this on WebWalkerQA and xBench-DS, which are 100% held out and where MemAgent leads all methods. Within GAIA, MemAgent also gains more on held-out L2/L3 (+11.6 pp, 51.8% → 63.4%) than on L1 (+9.4 pp, 73.6% → 83.0%), showing that the gains come from cross-task transfer rather than L1 replay.
Q4. "Insufficient statistical validation and cost accounting. The main results are reported as single accuracy numbers, without multiple runs, confidence intervals, or significance tests. ..."
A4. The task-level gains across the suite are large and statistically significant:
| Tasks solved | All 435 tasks | Held-out 382 tasks |
|---|---|---|
| No Memory | 284 | 245 |
| Random Routing | 290 | 254 |
| MemAgent w/o training | 295 | 257 |
| MemAgent | 328 | 284 |
MemAgent solves 44, 38, and 33 more tasks than these baselines on all 435 tasks (p = 0.001, 0.005, 0.013), and 39, 30, and 27 more on the 382 held-out tasks (p = 0.002, 0.017, 0.032). MemAgent also beats the hindsight-best single provider on each benchmark (AWM on GAIA, ExpeL on WebWalkerQA, SkillWeaver on xBench-DS), while every fixed provider trails MemAgent by at least 7.0 pp on some benchmark, a clear and consistent advantage. For single-run reporting, we follow MemEvolve [1] on the same three benchmarks, whose Table 3 likewise reports a single pass@1 accuracy per method. One pass over the 435 tasks already uses over 100M tokens and 10,000 paid search calls.
Q5. "Since all 13 long-term memory providers require LLM calls during storage, the authors should clarify whether the reported overhead of less than 0.3% excludes memory construction, trajectory summarization, provider updates, and training costs."
A5. The ~1.2 s per task (<0.3%) in Table 5 is router decision latency. To account for the full end-to-end cost per task, we tallied all LLM calls across the task lifecycle (task agent, retrieval, END trajectory summary, and storage in the 3–8 selected providers) on our GAIA logs:
| Per GAIA task | Tokens | API calls |
|---|---|---|
| No Memory | 0.17M | 13.6 |
| Random Routing | 0.27M | 28.8 |
| PromptRoute | 0.28M | 27.8 |
| MemAgent | 0.30–0.32M | 30 |
MemAgent updates only the 3–8 providers chosen at END rather than all 13. Storage runs after the final answer and is shared across later tasks, and router training is a one-time offline cost. The revised version retitles Table 5 "Routing overhead" and adds this table.
We hope these clarifications address the reviewer's concerns and would be grateful if the reviewer would reconsider the score.
[1] MemEvolve: Meta-Evolution of Agent Memory Systems. ICML 2026.
[2] Memory-R1: Enhancing Large Language Model Agents to Manage and Utilize Memories via Reinforcement Learning. ACL 2026.
[3] MemRL: Self-Evolving Agents via Runtime Reinforcement Learning on Episodic Memory. ArXiv 2026.
Polite Follow-up as the Discussion Period Concludes
Dear Reviewer DQ8M,
With the discussion deadline approaching, we wanted to politely check whether our response and revised manuscript have resolved your three core concerns: (1) clarifying our methodological contribution relative to single-store Memory-R1/MemRL and MemEvolve's LightWeight, (2) auditing the BEGIN memories and reporting strict held-out results excluding GAIA L1 across 382 disjoint tasks, and (3) providing full end-to-end storage and LLM cost accounting in Table 5b.
Please let us know if anything remains unclear before the discussion period ends. If our clarifications and revisions have addressed your concerns, we would deeply appreciate it if you could reconsider your rating. Thank you again for your time and helpful feedback.
Thank you for the detailed response. The results excluding GAIA L1 alleviate my concerns. I will keep my score unchanged for now and discuss the paper’s core contribution and applicability with the other reviewers in the next phase. I remain open to revising my score based on that discussion.
Response to Reviewer DQ8M (1/2)
We thank the reviewer for recognizing the practical problem, the three-phase design, and the clarity of the paper.
A1. Existing work, including adaptive methods such as MemEvolve [1], designs a single memory system. MemAgent addresses an orthogonal problem: learning how to route across a set of heterogeneous memories, so any new memory design can plug in as a provider. Specifically:
Simple combinations of the 13 providers fall well short of MemAgent (75.7% avg.): concatenating all retrieved memories gets 69.8% (All-Concat) and fusing them with an LLM gets 71.0% (All-Fuse). On the same pool, MemAgent also beats Random Routing (67.2%), the untrained router (68.1%), and a GPT-5-mini router prompted with rules mined from our training data (PromptRoute, 70.9%).
A2. Memory-R1 [2] and MemRL [3] manage entries inside a single memory store, which is complementary to routing across multiple stores. Memory-R1 trains an RL policy to apply ADD/UPDATE/DELETE/NOOP to a single dialogue memory bank, and MemRL learns utility scores to re-rank entries inside a single episodic store. MemAgent instead learns which heterogeneous memories to read from and write to, and either method could be added to our pool as a provider. For adaptive memory design, Table 1 already compares against the closest work: LightWeight, the memory system evolved and released by MemEvolve [1]. MemAgent beats LightWeight by +7.1 pp on average (75.7% vs. 68.6%). We have added a discussion of Memory-R1 and MemRL to Related Work.