Abstract: With advances in CLIP-based models for long-text processing, the insufficient ability to capture visual details has emerged as a key bottleneck for fine-grained image–text matching. Our ...

Penguin-VL is a compact vision-language model family built to study how far multimodal efficiency can be pushed by redesigning the vision encoder, rather than only scaling data or model size.

Abstract: Dead Reckoning (DR) algorithms using IMU and wheel encoders can continuously provide positioning information in scenarios where GNSS signals are lost, providing strong autonomy and playing a ...

Conventional dual fuel heat pumps lack the intelligent control mechanisms to efficiently manage the switch between heat pump and furnace, leading to sub-optimal energy usage and, in some cases, ...