When Stored Evidence Stops Being Usable: Scale-Conditioned Evaluation of Agent Memory
cs.AI
Submitted: 2026-05-08
Updated: 2026-10-29
Comments: Accepted at NeurIPS 2026, Evaluations and Datasets Track (poster). Updated to the camera-ready version. 25 pages, including references and appendices
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Toolformer: Language Models Can Teach Themselves to Use Tools
- ReAct: Synergizing Reasoning and Acting in Language Models
- AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation
- WebArena: A Realistic Web Environment for Building Autonomous Agents
- MemGPT: Towards LLMs as Operating Systems
- MemoryBank: Enhancing Large Language Models with Long-Term Memory
- A-MEM: Agentic Memory for LLM Agents
- MemOS: An Operating System for Memory-Augmented Generation (MAG) in Large Language Models
- LightMem: Lightweight and Efficient Memory-Augmented Generation
- HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models
- LiCoMemory: Lightweight and Cognitive Agentic Memory for Efficient Long-Term Reasoning
- LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory
- Evaluating Very Long-Term Conversational Memory of LLM Agents
- MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Tasks
- Memory in the Age of AI Agents
- RULER: What's the Real Context Size of Your Long-Context Language Models?
- Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactions
- Anatomy of Agentic Memory: Taxonomy and Empirical Analysis of Evaluation and System Limitations
- Qwen3 Technical Report
- gpt-oss-120b & gpt-oss-20b Model Card
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents