ManiGuard: Evaluated VLA Checkpoints Collection Every checkpoint behind the paper's rollout evaluations: 4 SFT policies × 6 families, with the scaling / prompt-format / sim-to-real studies. • 54 items • Updated Aug 18 • 2
ManiGuard: Benchmark & Datasets Collection ManiGuard-Bench tasks, the long-finger Franka asset, and the 6-family SFT demo trajectories (LeRobot v2.1). • 9 items • Updated Aug 20 • 3
IDEAS-Lab-Northwestern/datagen-clutter-v1-joint-5cam Franka Panda • Updated Aug 20 • 2.2k episodes • 848
IDEAS-Lab-Northwestern/datagen-cabinet-v1-joint-5cam Franka Panda • Updated Aug 20 • 1.4k episodes • 1.54k
view article Article Red Teaming with RL: Exploiting Tinker API for Harmful RL on 235B Model georgefen • Jan 1 • 19