Communications of the ACM

Building Intelligent Agents with Neuro-Symbolic Concepts

The agent acquires a vocabulary of neuro-symbolic concepts for objects, relations, and actions, represented through a ...
GitHub

MELLE: Autoregressive Speech Synthesis without Vector Quantization

An unofficial PyTorch implementation of "Autoregressive Speech Synthesis without Vector Quantization" paper. This repository provides a complete pipeline for training and inference of the MELLE model.
IEEE

Multirate Neural Image Compression with Adaptive Lattice Vector Quantization

Abstract: Recent research has explored integrating lattice vector quantization (LVQ) into learned image compression models. Due to its more efficient Voronoi covering of vector space than scalar ...
GitHub

VQ-VLA: Improving Vision-Language-Action Models via Scaling Vector-Quantized Action Tokenizers

VQ-VLA is an innovative vector quantization based action tokenizer built upon the largest-scale action trajectory dataset to date, leveraging over 100 times more data than previous approaches. It ...