About
Research & Development
CoMoE: Making Large Language Models Efficient on Edge Devices
TLDR: CoMoE is a new framework that combines expert aggregation and offloading to efficiently deploy large Mixture-of-Experts (MoE) language models on resource-constrained edge devices....
Research & Development
Advancing Handwritten Mathematical Expression Recognition with Large-Scale Data and Models
TLDR: Researchers developed Tex80M, the largest dataset of 80 million mathematical formulas, generated synthetically from LaTeX. Using this, they trained TexTeller, the first large-scale...
Research & Development
Forecasting Urban Bus Routes with Advanced Deep Learning
TLDR: GSMT is a novel deep learning model designed for accurate multi-bus trajectory prediction in urban environments. It combines Graph Attention Networks and Recurrent...
Research & Development
Fake-Mamba: A New Approach to Real-Time Speech Deepfake Detection
TLDR: Fake-Mamba is a novel system for detecting synthetic speech in real-time. It utilizes a bidirectional Mamba architecture with an XLSR front-end, offering significant...
Research & Development
TEN: A Neurosymbolic Approach to Extracting Tables from Unstructured Text
TLDR: TEN is a new neurosymbolic system that extracts tables from messy text by combining a large language model (LLM) with a symbolic checker....
Research & Development
Automating Prompt Creation for Better Text Correction and Simplification
TLDR: APIO is a novel method that automatically creates and refines prompts for large language models, achieving state-of-the-art performance in Grammatical Error Correction and...