3 papers
cs.CV2026
OpenGlass: Ultra-Low-Power On-Device AI Eyewear with Event-based Vision
Pietro Bonazzi, Julian Moosmann, Ahmet Celik +2
Smart eyewear enables unobtrusive, context-aware interaction through multimodal sensors and on-device intelligence, but is severely limited by power, memory, and compute constraint…
cs.LG2026
TyphoonMLA: A Mixed Naive-Absorb MLA Kernel For Shared Prefix
Ahmet Caner Yüzügüler, Ahmet Ãelik, Jiawei Zhuang +1
Multi-Head Latent Attention (MLA) is a recent attention mechanism adopted in state-of-the-art LLMs such as DeepSeek-v3 and Kimi K2. Thanks to its novel formulation, MLA allows two…
cs.LG2026
SINQ: Sinkhorn-Normalized Quantization for Calibration-Free Low-Precision LLM Weights
Lorenz K. Müller, Philippe Bich, Jiawei Zhuang +3
Post-training quantization has emerged as the most widely used strategy for deploying large language models at low precision. Still, current methods show perplexity degradation at…