MemoWM: How World Models Change What Agents Need to Remember

arXiv:2610.10778 · cs.LG, cs.AI · Submitted 2026-10-07 · Read on arXiv

cs.LG, cs.AI

Submitted: 2026-10-07

Updated: 2026-10-07

Code: https://github.com/Feld-maxiu/MemoWM

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• World Model — 1× in this paper · explained in 7 episodes

Sources

• V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning
• Agentic Learner with Grow-and-Refine Multimodal Semantic Memory
• Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory
• M2A: Multimodal Memory Agent with Dual-Layer Hybrid Memory for Long-Term Personalized Interactions
• MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory
• AUGUSTUS: An LLM-Driven Multimodal Agent System with Contextualized User Memory
• WorldMemArena: Evaluating Multimodal Agent Memory Through Action-World Interaction
• Omni-SimpleMem: Autoresearch-Guided Discovery of Lifelong Multimodal Agent Memory
• SimpleMem: Efficient Lifelong Memory for LLM Agents
• Agent Memory: Characterization and System Implications of Stateful Long-Horizon Workloads
• MemGPT: Towards LLMs as Operating Systems
• Zep: A Temporal Knowledge Graph Architecture for Agent Memory
• When Does Belief-Based Agent Memory Help? Reliability-Conditional Updating and Provenance-Capped Poisoning Defense
• D-MEM: Dopamine-Gated Agentic Memory via Reward Prediction Error Routing
• Voyager: An Open-Ended Embodied Agent with Large Language Models
• WebQuest: A Benchmark for Multimodal QA on Web Page Sequences
• MIRIX: Multi-Agent Memory System for LLM-Based Agents
• MemWM: Memory-Augmented Text-Based World Model
• LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory
• LongMemEval-V2: Evaluating Long-Term Agent Memory Toward Experienced Colleagues

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks