Memory Management for Agents
Methods for managing memory in LLM agents, including consolidation, compression, and learnable memory skills under budget constraints.
37 papers · 3 months
Where this stands
The written synthesis of this thread is for subscribers. Subscribe.
Results across this thread
2 reported results from the papers in this thread.
The table is for subscribers. Subscribe to see every reported number side by side.
How this thread developed
June 2026 · University of Science and Technology of China, Microsoft
Memory as state management instead of semantic retrieval improves long-horizon agents
Proposes a memory system based on execution-state tree, outperforming semantic retrieval.
June 2026 · Stanford University, Independent Researcher
First systematic characterization of agent memory systems across long-horizon tasks
Provides the first systematic characterization and taxonomy of agent memory systems.
June 2026 · Carnegie Mellon University, University of California, Berkeley
Shared agent memories boost performance without coordination
Shared repository for agent-generated trajectories boosts performance without explicit coordination.
June 2026 · Johns Hopkins University, Apple
Letting language models decide when to compact their context windows
SelfCompact lets agents dynamically decide when to compact context using summarization and rubrics.
July 2026 · Stanford University
Treating memory management as a trainable cognitive skill boosts LLM performance 2x–4x in long-horizon tasks
Memory management as a trainable skill improves long-horizon performance.
July 2026
Memory consolidation outperforms retention under tight budgets, but retention wins when budgets are loose.
Formal trade-off between memory consolidation and retention under different budget regimes.
23 further papers
Sept 2026 · Zhejiang University, Tencent
Program repair improves when memories match repository needs and repair stages
Introduces an adaptive memory-retrieval framework that tailors memory to repository needs and repair stages, advancing memory management for code repair agents.
Sept 2026 · Massachusetts Institute of Technology, Carnegie Mellon University
Saliency-driven workspace tokens give robots lightweight task memory
Introduces workspace tokens as a lightweight latent memory mechanism for robots, trained via VLM distillation to capture task-salient historical info.
Sept 2026
Adaptive memory graphs improve collaboration in multi-agent language-model systems
Graph-structured memory representation of prerequisites, actions, and outputs for agent systems.
Sept 2026 · National University of Defense Technology
Self-configured memory views improve retrieval across long conversations
Introduces self-configured memory views that structure retrieval across long conversations, advancing memory organization techniques.
Sept 2026 · Mem0
DolphinBench measures agent memory through tasks, cost, and latency
Benchmarks long-term memory via agent task completion, covering cost, latency, and real-world personas.
Sept 2026 · Zhejiang University, Alibaba Group
Checkpoint testing reveals hidden weaknesses in self-evolving agent memories
Introduces checkpoint-based evaluation revealing hidden weaknesses in self-evolving agent memories.
Sept 2026 · NLPR&MAIS, Institute of Automation, Chinese Academy of Sciences
Bounded visual workspaces improve multimodal agents’ accuracy and efficiency
Stores visual artifacts in a ledger as an alternative to appending to context, a memory management technique.
Sept 2026 · Huzhou Normal University, Alibaba Group
C3M preserves multimodal evidence across sessions within fixed memory budgets
Introduces C3M, a memory architecture that preserves multimodal evidence across sessions under fixed storage and retrieval budgets.
released code
14 of 37 papers shown