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: This paper presents a fully reconfigurable HDR intensity/event hybrid SPAD sensor. The fully reconfigurable scheme enables both pixel types to be freely blended and operated simultaneously.

In vision-language models (VLMs), visual tokens usually consume a significant amount of computational overhead, despite their sparser information density compared to text tokens. To address this, ...