APFT fine-tuning reduces OpenVLA failure under attention-hijacking patches from 100% to 25.9% in simulation and raises real-world success from 23.0% to 67.4%.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
VLAGuard: A Framework for Evaluating and Mitigating Physical Attention Hijacking in Vision-Language-Action Robots within Wireless Sensor Networks
APFT fine-tuning reduces OpenVLA failure under attention-hijacking patches from 100% to 25.9% in simulation and raises real-world success from 23.0% to 67.4%.