Santa Fe Institute

For computational devices, talk isn't cheap

Every task we perform on a computer requires different components of the machine to interact with one another — to ...
IEEE

Efficient Encoding and Decoding of Voxelized Models for Machine Learning-Based Applications

Abstract: Point clouds have become a popular training data for many practical applications of machine learning in the fields of environmental modeling and precision agriculture. In order to reduce ...

Orchard Therapeutics Celebrates Addition of Metachromatic Leukodystrophy to the U.S. Recommended Uniform Screening Panel

Newborn screening is the only practical means of diagnosing MLD prior to the onset of symptoms, which is key to achieving ...
GitHub

PersPose: 3D Human Pose Estimation with Perspective Encoding and Perspective Rotation

This is the offical Pytorch implementation of PersPose, which estimates the 3D positions of joints from individual images. Below is the overall framework. We design Perspective Encoding (PE) to encode ...
Electrical and Computer Engineering - Carnegie Mellon University

The Future of AI Compression Was Hiding In Your Video Player

Researchers discover that video compression technology is also great at compressing AI model data, earning Micro 25 Best Paper Award.
IEEE

End-to-End Learning System for Symbol Decoding in Optical Communication

Abstract: In this paper we propose an autoencoder-based End-to-End (E2E) Deep Learning for symbol decoding in fiber optical communications systems. The autoencoder was trained with data encoded in two ...
GitHub

Free Draft-and-Verification: Toward Lossless Parallel Decoding for Diffusion Large Language Models

We propose FreeDave (Free Draft-and-Verification), a fast sampling algorithm for diffusion language models, which achieves lossless parallel decoding via a pipeline of parallel-decoded candidate ...