Firehose

Filtered to Papers, tagged “deep learning inference” · clear filters

Browse: People · Companies · Papers · Podcasts · Hacker News · Deep dives

All PeopleCompaniesPapersPodcastsHacker News

Browse by tag

continual learning 26reinforcement learning 9language models 5large language models 4robotics 4agents 3generative models 3multi-agent systems 3recursive self-improvement 3vision-language models 3
16 SEP 2026 · Paper

The Other Half of the Memory Wall: Serving 35B MoEs from SSD with Trained Routing Prediction

This paper presents a new technique to reduce memory usage and speed up inference for large neural networks, allowing them to run on consumer hardware with limited memory. A practitioner might care about this because it enables the deployment of large models in edge devices and reduces the need for expensive storage.

mixture-of-expertsmemory efficiencydeep learning inferenceneural network pruningsparse models