EMPIRIC: Experiment-Driven Learning of Residual World Models for Robot Planning
cs.RO, cs.AI, cs.LG
Submitted: 2026-09-28
Updated: 2026-09-28
Project page: https://yichao-liang.github.io/empiric
Terminology
Sources
- Bayesian Online Changepoint Detection
- Combining Physical Simulators and Object-Based Networks for Control
- ARC-AGI-3: A New Challenge for Frontier Agentic Intelligence
- From Pixels to Predicates: Learning Symbolic World Models via Pretrained Vision-Language Models
- Agentic Real2Sim: Physics-based World Modeling with Vision-Language Agents
- Generating Code World Models with Large Language Models Guided by Monte Carlo Tree Search
- CaP-X: A Framework for Benchmarking and Improving Coding Agents for Robot Manipulation
- NeuralSim: Augmenting Differentiable Simulators with Neural Networks
- Bayesian Active Learning for Classification and Preference Learning
- Code World Models for General Game Playing
- ASPIRE: Agentic /Skills Discovery for Robotics
- SimFoundry: Modular and Automated Scene Generation for Policy Learning and Evaluation
- SAM 2: Segment Anything in Images and Videos
- Learning POMDP World Models from Observations with Language-Model Priors
- ENPIRE: Agentic Robot Policy Self-Improvement in the Real World
- One-Shot Real-to-Sim via End-to-End Differentiable Simulation and Rendering
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
- HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
- APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies
- Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving