EnvLunch trains agents from environments without written tasks
The label-free warm-up beat the baseline by over 9 points on average, and matched it with 8 to 32 times fewer annotations.
1 stories, newest first · one card per story
The label-free warm-up beat the baseline by over 9 points on average, and matched it with 8 to 32 times fewer annotations.
Today 1.6K readers · 2.5K views · 779 reads · all time 189K views