b:botstest.Submit a resource ↗

Benchmarks / Research

libero camera ablation · chaoshengsc

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

diffusion-policyimitation-learninglerobotliberomujoco
Publisher repository ↗
libero camera ablation · chaoshengsc resource page screenshot
Resource screenshot · captured 2026-10-08 · visually reviewed

Overview

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

Publisher summary: “Camera-input ablations of π0, π0.5, OpenVLA and SmolVLA on LIBERO-Spatial (blackout, frozen frame, noise, single camera), plus reproduction of five public robot-policy checkpoints in…”

Use its repository to inspect the implementation, documentation and supported workflows. This directory record identifies the maintainer, code license and related resources; it does not claim that the project has been installed or tested on your robot.

Project information

Maintainer
chaoshengsc
Primary language
Python
Repository code license
MIT
Model-weight license
Not recorded
Dataset license
Not recorded
Hardware design license
Not recorded
Maintenance
Not archived in the last recorded metadata
Last observed repository push
2026-10-07

Code licenses do not automatically cover model weights, datasets or hardware designs. Unrecorded license details require confirmation with the publisher.

Documentation sections

  • Camera-input ablations of four robot policies on LIBERO-Spatial
  • 1. Camera ablations
  • 1.1 All cameras black
  • 1.2 One camera, noise, frozen image
  • 1.3 Where blind π0 succeeds, it follows the sighted path
  • Notes on these numbers

Browse this resource’s categories

Explore related collections

Browse Benchmarks →