Jack Monas commited on
Commit
cd32b64
·
1 Parent(s): 11f084a
Files changed (1) hide show
  1. app.py +6 -2
app.py CHANGED
@@ -86,8 +86,8 @@ def main():
86
  st.markdown("## Motivation")
87
  st.write(
88
  "Real-world robotics faces a fundamental challenge: environments are dynamic and change over time, "
89
- "making consistent evaluation of robot performance difficult. World models offer a solution by "
90
- "learning to simulate complex real-world interactions from raw sensor data. We believe these learned simulators will enable "
91
  "robust evaluation and iterative improvement of robot policies without the constraints of a physical testbed."
92
  )
93
  st.image(
@@ -99,6 +99,10 @@ def main():
99
 
100
  st.markdown("## The Challenges")
101
 
 
 
 
 
102
  st.markdown("#### Compression Challenge")
103
  st.write(
104
  "In the Compression Challenge, your task is to train a model to compress our robots logs effectively while preserving the critical details needed to understand and predict future interactions. Success in this challenge is measured by the loss of your model—the lower the loss, the better your model captures the complexities of real-world robot behavior."
 
86
  st.markdown("## Motivation")
87
  st.write(
88
  "Real-world robotics faces a fundamental challenge: environments are dynamic and change over time, "
89
+ "making consistent evaluation of policy performance difficult. Driven by recent advances in video generation, World models offer a solution by "
90
+ "learning to simulate complex real-world interactions from image tokens paired with action data. We believe these learned simulators will enable "
91
  "robust evaluation and iterative improvement of robot policies without the constraints of a physical testbed."
92
  )
93
  st.image(
 
99
 
100
  st.markdown("## The Challenges")
101
 
102
+ st.write(
103
+ "To accelerate reseach in this area, we're launching the 1X World Model Challenge. In this competition, you'll predict future first-person observations of the EVE Android using over 100 hours of video and action data with the ultimate goal of solving policy evaluation."
104
+ )
105
+
106
  st.markdown("#### Compression Challenge")
107
  st.write(
108
  "In the Compression Challenge, your task is to train a model to compress our robots logs effectively while preserving the critical details needed to understand and predict future interactions. Success in this challenge is measured by the loss of your model—the lower the loss, the better your model captures the complexities of real-world robot behavior."