Policy evaluation

Research

Simulation environments for evaluating manipulation policies.

Research

One framework to evaluate any VLA model on any robot simulation benchmark.

Research

ActProbe provides probes and benchmarks for early failure detection in generative robot policies.

Research

Code for "Evaluating Robot Policies in a World Model".

Research

Benchmarking suite to evaluate 🤖 robotics computing performance. Vendor-neutral. ⚪Grey-box and ⚫Black-box approaches.

Research

Faultline studies LIBERO policy robustness using perturbation and ablation workflows.

Research

Isaac Lab - Arena is a robotics simulation framework that enhances NVIDIA Isaac Lab by providing a composable, scalable system for creating diverse simulation environments and evaluating robot…

Research

LIBERO Camera Ablation studies how camera-input changes affect robot-policy behavior.

Research

OceanGym: A Benchmark Environment for Underwater Embodied Agents

Research

Open-source robotics simulator for coding agents: HTTP/JSON + MCP control, Newton physics, wgpu rendering, ROS 2, and reproducible benchmarks.

Research

ParcelStow evaluates imitation-learned robot-policy robustness under execution-timing variation in Isaac Lab.

Research

Pre-Execution Return studies robot initial-state robustness around frozen VLA policies.

Research

A research benchmark for robot manipulation policy evaluation.

Research

Repository associated with paper titled "RoboGaze: Evaluating Robot World Models via Structured Vision-Language Analysis"

Research

A simulation benchmarking platform for evaluating generalist robot policies

Research

RoboPianist Cost-Aware Intervention provides a benchmark for teacher interventions in RoboPianist.

Research

Benchmarking Coding Agents for Interactive World Modeling from Embodied Videos - Rethinking Embodied Real2Sim from a Software Engineering Perspective

Research

VLA Grounding Probe investigates SmolVLA language grounding using instruction perturbations.

Research

VLA Manipulation Ablations studies manipulation generalization and OpenVLA under constrained GPU memory.