EVOLVE-VLA: Test-Time Training from Environment Feedback for Vision-Language-Action Models

1 Open Issue Need Help Last updated: Dec 18, 2025

Open Issues Need Help

View All on GitHub

AI Summary: The user is asking for clarification regarding the number of test-time online rollout trajectories utilized by EVOLVE-VLA for fine-tuning a single task or suite. They specifically reference Figure 1(d) from the paper as an example to understand the experimental setup.

Complexity: 1/5
good first issue

EVOLVE-VLA: Test-Time Training from Environment Feedback for Vision-Language-Action Models

HTML