- Rating
- 2
- Confidence
- 4
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper introduces Librarian Agents, a framework for anticipatory memory construction over long document corpora. Instead of performing retrieval independently for each query, a builder agent proactively reorganizes the corpus into a hierarchical filesystem containing synthetic summaries, merged artifacts, and provenance links to the original documents. A separate answering agent then navigates this filesystem to answer future queries.
Critical strengths
- A new formulation of anticipatory memory construction. The paper identifies an important limitation of reactive retrieval systems: they repeatedly rediscover corpus-level structure for each query. Modeling reusable memory construction as an MDP optimized for downstream answering performance is a compelling and potentially general framework.
- Strong empirical gains across multiple settings. The RL-trained builder substantially outperforms the reported retrieval, graph, autocompaction, and non-RL builder baselines on BrowseComp-Plus and also transfers to BRIGHT domains. The provenance-based fallback mechanism is particularly useful because it allows the answering agent to recover information omitted from synthetic summaries.
Critical weaknesses
- Insufficiently specified baseline comparisons. Although the reported gains are substantial, the comparison protocol does not provide enough detail to establish whether the baselines are comparably configured. Appendix A.7 states that source corpora, answering models, answer-generation budgets, and scoring procedures are matched, but retrieval settings, context allocation, tool access, and baseline tuning remain insufficiently documented. This makes it difficult to attribute the improvements specifically to the proposed memory-construction approach. The training-set size is also inconsistent: Section 3 and Appendix A.7 report 680 tasks, whereas Appendix A.1 reports 780.
- Limited evidence for practical efficiency gains. Librarian Agents incurs approximately 179K construction tokens per corpus and substantially increases answer-time token consumption relative to the base builder. The reported efficiency advantage is based on tokens per expected correct answer, rather than lower raw token consumption, and the reuse analysis primarily compares against the non-RL builder. Consequently, it does not establish when the proposed approach is more economical than strong retrieval baselines. A comparison of accuracy and total cost across query volumes, including clearly defined crossover points, is needed to substantiate the practical efficiency claims.
- The incremental novelty is not sufficiently established. Query-independent precomputation and amortization have been explored in Sleep-time Compute, hierarchical corpus summarization in RAPTOR, and outcome-driven RL for memory management in Memory-R1. The potentially distinctive contribution is their integration into a learned, reusable filesystem over external corpora. However, the paper lacks direct comparisons with these closely related approaches. Moreover, the RL builder produces and exposes substantially more synthetic content than the base builder, leaving unclear whether the gains arise from better organization or increased coverage and computation. Comparisons at matched memory sizes and token budgets, together with hierarchy ablations, would help establish the contribution.
Questions for authors
- Please clarify the BrowseComp-Plus data split. Section 3 and Appendix A.7 report 680 training tasks and 50 held-out evaluation tasks, whereas Appendix A.1 reports 780 training tasks. Which split was used for the reported results?
- Appendix A.7 states that source corpora, answering models, answer-generation budgets, and scoring procedures are matched. Could you additionally specify each baseline’s retrieval/context limits, tool access, tuning procedure, and construction-time and answer-time token consumption? These details would help determine whether the gains reflect improved memory construction or differences in available computation.
- Could you clarify the direction and interpretation of the crossover at approximately 31 questions per corpus? The paper reports an accuracy-normalized cost advantage at ten questions while raw per-query token consumption remains higher. Over which query-volume range is Librarian Agents more economical than the base builder, and how does this comparison change when using the strongest retrieval baselines?
- How was the reliability of the Gemini-based correctness judge assessed, particularly for BRIGHT-derived answers? Could you provide agreement with human judgments on a representative sample, or an assessment using an independent judge, to establish whether the reported gains are robust to the scoring procedure?
- How does the proposed approach differ substantively from Sleep-time Compute, RAPTOR, and Memory-R1 beyond their integration into a filesystem interface? Do you have comparisons or ablations at matched memory sizes and token budgets that distinguish the benefit of learned hierarchical organization from increased synthetic-content coverage?
7 comments
Response to Reviewer iHq6 [2/N]
Clarifying Novelty
Librarian Agents introduces anticipatory workspace construction as a learned capability: a builder transforms a corpus into a persistent, navigable filesystem before receiving downstream queries, and is trained through reinforcement learning on the answering agent’s subsequent performance. There are distinct differences from the Sleep-time Compute, RAPTOR, and Memory-R1 prior work. We will discuss each individually.
Sleep-time Compute performs unstructured chain-of-thought precomputation over short contexts. Notably, this computation isn’t a learned behavior but elicited via prompting. In contrast, in Librarian Agents we instead formulate an objective to learn how to build a persistent, navigable workspace in the form of a multi-document filesystem with explicit reference links over large corpora and formulate this as an MDP. We can view the autocompaction baseline as an instantiation of the sleep-time compute paper, which we outperform. RAPTOR relies on fixed, heuristic bottom-up embedding clustering and static summarization prompts; it is not optimized end-to-end for downstream agentic navigation, and its lossy tree compression often smooths over fine-grained cross-document relational links. We think that approaches like Raptor can be complementary to Librarian Agents as we state in Section 5, where the clustering of the builder in RAPTOR can be a good structured filesystem to further optimize with the RL fine-tuned builder from Librarian Agents. We leave this analysis to future work.
Memory-R1 studies memory management and utilization for multi-session dialogue. It incrementally takes dialogue history and maintains an external bank of extracted conversational facts using the Mem0 harness, training both a memory manager and an answering agent with outcome-driven RL, using standard policy-gradient algorithms such as PPO and GRPO. Librarian Agents instead studies upfront transformation of a document corpus into a reusable hierarchical workspace of synthetic files and source references, before downstream queries are observed. Though both methods create textual memories, a contribution here is that the agent organizes this information into an accessible data structure for future reuse. Furthermore, our training formulation not only relies on terminal answer correctness but also leverages periodic probes of improvements in answerability for better credit assignment during training. The key distinction is the scope and organization of memory construction: Memory-R1 learns incremental maintenance and retrieval-based use of conversational memories, whereas Librarian Agents learns corpus-level synthesis and organization intended to shift useful computation ahead of query time and amortize it across future queries.
To summarize, we identify three distinct components of memory building that these methods explore:
| Built before queries | Structured, navigable memory | Builder learned from downstream reward | |
|---|---|---|---|
| Sleep-time Compute | Yes | No | No |
| RAPTOR | Yes | Yes | No |
| Memory-R1 | No | No | Yes |
| Librarian Agents | Yes | Yes | Yes |
The three properties depend on each other. Building before queries is what allows cost to be amortized, but it means the builder cannot know what will be asked. Memory therefore has to stay navigable, so that an answering agent can find what it needs for a question nobody anticipated. And which content and organization will be useful is only revealed by downstream answers, so it has to be learned from that signal: with the same filesystem interface and the same construction budget, the untrained builder reaches 24.8% against 63.7% for Librarian, and RAPTOR's fixed structure reaches 21.4%.
Recent blogs from two industry labs, Engram [1] and Anthropic [2] (released after our paper's submission), independently isolate this capability of anticipation and "dreaming" as a core bottleneck for models at scale, finding that optimizing for hierarchical note creation improves both downstream performance and inference efficiency. This is a key problem which we begin to tackle with our Librarian Agents formulation.
Response to Reviewer iHq6 [3/N]
BrowseComp-Plus Data Split Clarification
All reported results use 680 training tasks, 50 validation tasks for hyperparameter/checkpoint selection, and 50 held-out evaluation tasks. The reference to “780 training tasks” incorrectly described the total dataset size as the training-set size. We will make the 680/50/50 split explicit and consistent throughout the manuscript.
Baseline configuration and reproducibility
For full reproducibility, we have released our full codebase in an anonymous repository below. Happy to share any details to make other details clear.
Anonymous URL: https://anonymous.4open.science/r/librarian-agents-anonymous-review-DB27/
Additional Legal-Domain Evaluation
We further evaluate Librarian Agents on a more realistic domain by training a Qwen 3.8-27B builder on the Harvey Legal Agent Bench [4] with the mini-swe-agent agent harness to provide our agents more flexibility with bash tools. On the benchmark, on the set of held out tasks in the validation set, training increased task-equal accuracy from 26.2% to 65.6%, placing the trained builder above all five baselines in our paper (the highest being GraphRAG at 59.15%); a frontier zero-shot Codex builder achieved 72.5%. Together with the prompt-optimization result above, this supports our claim that optimizing memory construction substantially improves a given builder, including frontier builders, rather than implying that a trained open-weight model must always outperform a much larger proprietary agent.
| Model / Method | Score |
|---|---|
| Trained RL Librarian | 65.65% |
| Untrained/base model | 26.24% |
| Embedding retrieval | 42.01% |
| Reranker | 45.84% |
| GraphRAG | 59.15% |
| AutoCompaction | 26.98% |
| RAPTOR | 55.00% |
| Frontier zero-shot builder | 72.53% |
We explicitly measure the gap between LibrarianAgents and baselines and report 95% boostrapped confidence intervals.
| Method | Improvement (pp) | 95% CI |
|---|---|---|
| Embedding | +23.6 | [11.7, 36.6] |
| Reranker | +19.8 | [7.5, 34.4] |
| GraphRAG | +6.5 | [1.1, 14.6] |
| AutoCompaction | +38.7 | [30.3, 47.2] |
| RAPTOR | +10.6 | [0.4, 21.5] |
Please let us know if you find your concerns addressed, and if so, we would be grateful if you are willing to raise your score.
References
- Engram, Learning in a World of Notes (https://engram.com/blog/learning-in-a-world-of-notes)
- Anthropic, Components of Agent Memory (https://www.anthropic.com/research/components-of-agent-memory)
- Harvey AI, Legal Agent Bench: Law Firm Knowledge (https://www.harvey.ai/blog/legal-agent-bench-law-firm-knowledge)
Response to Reviewer iHq6 [1/N]
We thank Reviewer iHq6 for highlighting the novelty of modeling anticipatory memory construction as an MDP alongside our strong empirical results and provenance fallback mechanism. Below, we address each of your questions with new compute-matched experiments, baseline specifications, cost crossover analyses, and cross-judge evaluations.
Following your suggestion, we evaluated RAPTOR and an expanded-budget untuned builder on the amortization QA benchmark in Table 2:
Increasing the untuned builder’s construction budget improves accuracy by only 0.6 percentage points, whereas Librarian exceeds this token-matched control by 38.9 points. Librarian also exceeds token-matched RAPTOR by 42.3 points. These comparisons support the conclusion that increasing the untuned builder’s construction budget alone does not reproduce the learned builder’s gains. We distinguish this conclusion from the stronger claim that hierarchy alone explains the improvement. Matching construction tokens does not automatically match final memory size, information coverage, or answer-time consumption. The learned policy can improve which information is synthesized and how it is organized.
Appendix A.5 provides complementary content-preserving interventions. In its separate matched run, we run an ablation over original, flattened, and shuffled memories:
Flattening therefore increases navigation cost by approximately 51% without lowering accuracy. Shuffling lowers point-estimate accuracy by 4.14 points at similar navigation cost. Hierarchy therefore reduces navigation cost, and semantic placement may also help accuracy, but organization alone does not explain the full gain. Note that this run uses a different question set, so its accuracies are not comparable to the 63.7% above.
The reviewer correctly highlights the distinction between raw per-query token expenditure and accuracy-normalized compute efficiency. Formally, let B_m represent the fixed corpus construction cost, A_m denote the average answer-time token consumption per query, and p_m denote downstream answer accuracy. Across q queries, the cumulative token expenditure is B_m + q A_m, yielding an accuracy-normalized cost per correct answer of (B_m + q A_m) / (q p_m). Because Librarian Agents achieve significantly higher accuracy (p_m), it yields better amortized efficiency per successful outcome within an empirical crossover window of 18–54 questions (centering near 31 questions per corpus). Furthermore, increasing the construction budget of an untuned base builder fails to yield competitive accuracy, as demonstrated in our token-matched baseline experiments. Thus, the observed performance advantage stems directly from end-to-end RL policy optimization rather than redundant token allocation.
We additionally compared efficiency against Rerank and Agentic RAG on the BrowseComp+ amortization cohort. Librarian used 8.81M measured tokens in total and 36.3K tokens per correct answer, compared with 17.40M and 74.0K for Rerank and at least 9.12M and 38.8K for Agentic RAG. The total-token crossover points are 3.6 questions per corpus against Rerank and 8.4 against Agentic RAG. Thus, at the observed amortization volume, Librarian achieves accuracy statistically comparable to both strong retrieval baselines while using fewer tokens.