LIBERO-MAX: Do Robot Policies Adapt When the World Changes?
cs.RO, cs.AI
Submitted: 2026-09-29
Updated: 2026-09-29
Comments: 42 pages, 15 figures. Project page: https://liberomax.github.io
Project page: https://liberomax.github.io/Geometry
License: http://creativecommons.org/licenses/by/4.0/
The gist: Robots must often continue a task after a target moves, the viewpoint shifts, or an obstacle appears, even though their earlier observations and committed actions reflect the previous scene.
Terminology
Abstract
Robots must often continue a task after a target moves, the viewpoint shifts, or an obstacle appears, even though their earlier observations and committed actions reflect the previous scene. Many simulation robustness benchmarks fix external conditions at reset, leaving this temporal challenge underexamined. We introduce LIBERO-MAX, a benchmark of 8,000 paired cases spanning eight types of changes to geometry, observations, appearance, clutter, and paths. Each pair compares task execution with and without a mid-task event, holding the task, initial state, policy seed, and pre-event action sequence fixed. This controlled comparison distinguishes event-associated regressions from failures already present without the change. Across fourteen current VLA, hybrid, and world-action policies, events reduce success by 11.0-25.7 percentage points. Event profiles reveal shared vulnerabilities to geometry and observation changes, while policy-family rankings interleave. Camera controls show that robustness reflects both competence under the changed conditions and the trajectory from which they are encountered; varying query cadence does not eliminate the gap. Together, the paired protocol and temporal diagnostics establish LIBERO-MAX as a reproducible testbed for diagnosing failures under mid-execution changes and measuring progress toward robot policies that remain effective as the world changes.
Sources
- Cosmos World Foundation Model Platform for Physical AI
- CausalWorld: A Robotic Manipulation Benchmark for Causal Structure and Transfer Learning
- V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning
- Motus: A Unified Latent Action World Model
- GR00T N1: An Open Foundation Model for Generalist Humanoid Robots
- RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control
- RT-1: Robotics Transformer for Real-World Control at Scale
- Genie: Generative Interactive Environments
- UniVLA: Learning to Act Anywhere with Task-centric Latent Actions
- Xiaomi-Robotics-0: An Open-Sourced Vision-Language-Action Model with Real-Time Execution
- WorldVLA: Towards Autoregressive Action World Model
- RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation
- Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models
- LIBERO-Safety: A Comprehensive Benchmark for Physical and Semantic Safety in Vision-Language-Action Models
- When Images Look Right and Retrieve Wrong: Coverage-Guided Cross-Scale Re-Indexing for Knowledge-Faithful Generative Perception
- MolmoAct2: Action Reasoning Models for Real-world Deployment
- Towards Generalizable Robotic Manipulation in Dynamic Environments
- From Intention to Execution: Probing the Generalization Boundaries of Vision-Language-Action Models
- LIBERO-Plus: In-depth Robustness Analysis of Vision-Language-Action Models
- Deep Visual Foresight for Planning Robot Motion
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