TL;DR: Turbolearn co-designs the switch ASIC and switch OS to split inference into a hardware fast path and a software normal path: the fast path performs lightweight feature extraction and classification at line rate, while only low-confidence flows are offloaded to the normal path for full inference.
Intelligent data plane (IDP) enables line-rate traffic analysis by embedding deep learning (DL) models in network switches. However, existing IDP systems must fit the entire inference pipeline into the switch data plane, forcing simplified models and sacrificing accuracy under ti…
Infrared vision-language models (IR-VLMs) extend thermal perception to open-vocabulary classification, image captioning, and visual question answering. However, their robustness to structured thermal perturbations and the stability of cross-modal semantic alignment remain insuffi…
This paper presents a review of the second LoViF Challenge on Real-World All-in-One Image Restoration. The challenge aims to advance unified image restoration under diverse real-world degradation conditions, including blur, low-light, haze, rain, and snow. It provides a common be…
Infrared remote-sensing imagery captures intensity structure, object-background contrast, and illumination-invariant cues often invisible in RGB imagery. Yet, most remote-sensing vision-language resources and models focus on visible-band semantics, leaving infrared vision-languag…
DiT video generation is latency-intensive due to iterative full-frame denoising, while prior cloud-edge methods largely rely on static inter-step decoupling and cannot leverage inter-frame similarity or adapt to system dynamics. We propose EcoVideo, an entropy-orchestrated framew…
Vision-Language-Action (VLA) models represent a promising direction for embodied intelligence in surgical robotics. Despite the prevalence of VLA benchmarks for general robotics, standardized evaluation platforms specifically designed for surgical contexts remain absent. To addre…