Showing cs.LGShow all
3 papers · 1 filter
cs.LG2026
Data Scarcity and Model Sparsity: Mixtures-of-Experts Overfit More to Repeated Data
Atindra Jha, Margaret Li, Jure Leskovec +2
As the supply of human-written text is exhausted, it has become standard practice to repeat language model training data. Prior work has studied data repetition for densely activat…
cs.LG2026
M*: A Modular, Extensible, Serving System for Multimodal Models
Atindra Jha, Naomi Sagan, Keisuke Kamahori +9
We are entering a new era of composite model architectures that integrate diverse components such as vision encoders, language backbones, diffusion and flow heads, audio codecs, ac…
cs.LG2026
VoxServe: Streaming-Centric Serving System for Speech Language Models
Keisuke Kamahori, Wei-Tzu Lee, Garv Ghai +5
Speech Language Models (SpeechLMs) are increasingly deployed for streaming speech generation, where audio playback must start quickly and never stall. Serving them requires running…