PonderPounce: A Pretrained MLLM as an Episode Context Engine for Robot Control
PONDERPOUNCE reuses an MLLM's native causal context and demonstrates real-world applicability on four tasks under asynchronous execution, with 60.98% mean success versus 40.67% for FrameSamp+Modul.
Suhwan Choi, Jaeyoon Jung, Sungkyung Kim et al.
· 0 citations