CrossBFM: Distilling a Shared Latent Behavior Space Across Humanoid Embodiments
日本語版は未提供のため、英語原文で表示しています。
要旨より
Behavior Foundation Models (BFMs) give humanoids a promptable policy over a latent behavior space, enabling one single vector to represent a motion to imitate, a pose to reach, or a reward to maximize. Forward-Backward representations successfully produce such spaces, but at the cost of hundreds of GPU-hours for a single robot.
要旨より。当社による要約は作成中です。