DeltaMem: Towards Agentic Memory Management via Reinforcement Learning
arXiv:2604.01560v1 Announce Type: new Abstract: Recent advances in persona-centric memory have revealed the powerful capability of multi-agent systems in managing persona memory, especially in conversational scenarios. However, these complex frameworks often suffer from information loss and are fragile across varying scenarios, resulting in suboptimal performance. In this paper, we propose DeltaMem, an agentic memory management system that formulates persona-centric memory management as an end-to-end task within a single-agent setting. To further improve the performance of our agentic memory manager, we draw inspiration from the evolution of human memory and synthesize a user-assistant dialogue dataset along with corresponding operation-level memory updating labels. Building on this, we in
View PDF HTML (experimental)
Abstract:Recent advances in persona-centric memory have revealed the powerful capability of multi-agent systems in managing persona memory, especially in conversational scenarios. However, these complex frameworks often suffer from information loss and are fragile across varying scenarios, resulting in suboptimal performance. In this paper, we propose DeltaMem, an agentic memory management system that formulates persona-centric memory management as an end-to-end task within a single-agent setting. To further improve the performance of our agentic memory manager, we draw inspiration from the evolution of human memory and synthesize a user-assistant dialogue dataset along with corresponding operation-level memory updating labels. Building on this, we introduce a novel Memory-based Levenshtein Distance to formalize the memory updating reward, and propose a tailored reinforcement learning framework to further enhance the management capabilities of DeltaMem. Extensive experiments show that both training-free and RL-trained DeltaMem outperform all product-level baselines across diverse long-term memory benchmarks, including LoCoMo, HaluMem, and PersonaMem.
Comments: preprint, under review
Subjects:
Computation and Language (cs.CL)
Cite as: arXiv:2604.01560 [cs.CL]
(or arXiv:2604.01560v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2604.01560
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Qi Zhang [view email] [v1] Thu, 2 Apr 2026 03:13:53 UTC (1,669 KB)
Sign in to highlight and annotate this article
Conversation starters
Daily AI Digest
Get the top 5 AI stories delivered to your inbox every morning.
More about
benchmarktrainingannounce
5 Patterns for Building Resilient Event-Driven Integrations
Point-to-point integrations are easy to build and easy to break. You wire up an API call from one system to another, it works in testing, and then a 30-second downstream outage in production causes a cascade of failures, lost state, and a manual cleanup effort that takes longer than the outage itself. Event-driven integration patterns address this directly. They decouple the systems involved so that no single failure propagates through the entire integration chain. The tradeoff is upfront design work, but the operational stability that results is not comparable to the alternative. Here are five patterns that appear in most well-built event-driven integrations, with examples of when and why each one matters. 1. Queue-Based Event Processing What it is: Instead of processing webhook events or
Choosing an AI Agent Orchestrator in 2026: A Practical Comparison
Running one AI coding agent is easy. Running three in parallel on the same codebase is where things get interesting — and where you need to make a tooling choice. There's no "best" orchestrator. There's the right one for your workflow. Here's an honest comparison of five approaches, with the tradeoffs I've seen after months of running multi-agent setups. The Options 1. Raw tmux Scripts What it is: Shell scripts that launch agents in tmux panes. DIY orchestration. Pros: Zero dependencies beyond tmux Full control over every detail No abstractions to fight You already know how it works Cons: No state management — you track everything manually No message routing between agents No test gating — agents declare "done" without verification Breaks when agents crash or hit context limits You become
How AI Is Changing the Way We Build Online Businesses
Not long ago, building an online business meant: months of development hiring developers large upfront costs Today? AI has completely changed the game. Now, one person can go from idea → to revenue faster than ever before. And this shift is just getting started. ⚠️ The Old Way vs The New Way Before AI: Build everything from scratch Spend weeks on infrastructure Launch slowly Iterate even slower With AI: Build faster Automate key tasks Launch quickly Iterate in real time The difference is massive. 🧠 AI Is Reducing the Cost of Building One of the biggest changes: 👉 Building is no longer the bottleneck AI helps with: generating content writing code automating workflows handling repetitive tasks What used to take weeks… 👉 now takes days ⚙️ Infrastructure Is No Longer the Hard Part Another s
Knowledge Map
Connected Articles — Knowledge Graph
This article is connected to other articles through shared AI topics and tags.
Discussion
Sign in to join the discussion
No comments yet — be the first to share your thoughts!