Memory Without Curation: Learning Multi-Stream Latent Memory for Robot Policies
cs.RO, cs.AI, cs.LG
Submitted: 2026-10-09
Updated: 2026-10-09
Project page: https://yilin-wu98.github.io/imprint_website/Abstract
Sources
- Cosmos 3: Omnimodal World Models for Physical AI
- GR00T N1: An Open Foundation Model for Generalist Humanoid Robots
- $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
- RoboDojo: A Unified Sim-and-Real Benchmark for Comprehensive Evaluation of Generalist Robot Manipulation Policies
- RMBench: Memory-Dependent Robotic Manipulation Benchmark with Insights into Policy Design
- Workspace Models: Lightweight Robotic Memory via Saliency-Driven Supervision
- MolmoAct2: Action Reasoning Models for Real-world Deployment
- Gated Memory Policy: In-Context Memorization and Adaptation
- $\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization
- RoboTTT: Context Scaling for Robot Policies
- BPP: Long-Context Robot Imitation Learning by Focusing on Key History Frames
- RoboCasa: Large-Scale Simulation of Everyday Tasks for Generalist Robots
- Scaling Short-Term Memory of Visuomotor Policies for Long-Horizon Tasks
- DINOv3
- A Careful Examination of Large Behavior Models for Multitask Dexterous Manipulation
- Learning Long-Context Diffusion Policies via Past-Token Prediction
- MEM: Multi-Scale Embodied Memory for Vision Language Action Models
- Wan: Open and Advanced Large-Scale Video Generative Models
- EventVLA: Event-Driven Visual Evidence Memory for Long-Horizon Vision-Language-Action Policies
Related papers
- FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
- MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
- RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
- ARC: A Reasoning Recipe for Robot Foundation Models
- HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
- APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies