About
Research & Development
Geometric Mixture Classifier: A Transparent and Efficient Approach to Multimodal Data Classification
TLDR: The Geometric Mixture Classifier (GMC) is a new discriminative model that addresses the challenge of classifying multimodal data by representing each class as...
Research & Development
Advancing Sentiment Analysis for Central Kurdish with BERT
TLDR: The paper introduces KuBERT, a BERT-based model for sentiment analysis in Central Kurdish, a low-resource language. It details the creation of a large...
Research & Development
Bridging the Data Gap: How Robots Learn with Limited and Incomplete Information
TLDR: Jingxi Xu's PhD research addresses the critical challenges of data sparsity and scarcity in robot learning. In tactile manipulation, methods like TANDEM, TANDEM3D,...
Research & Development
Precision Audio Alignment: Leveraging Cross-Attention and Confidence Weighting
TLDR: A new research paper introduces a novel method for multi-channel audio alignment that uses cross-attention mechanisms to model inter-channel dependencies and a confidence-weighted...
Research & Development
Enhanced Video Object Segmentation with SaSaSa2VA
TLDR: SaSaSa2VA is a novel method for referring video object segmentation (RVOS) that significantly improves upon previous models. It addresses limitations like sparse frame...
Research & Development
Trit-Planes: A Breakthrough in Low-Bit Quantization for LLMs
TLDR: PTQTP is a novel post-training quantization (PTQ) framework that compresses large language models (LLMs) to an ultra-low 1.58-bit representation using structured ternary {-1,0,1}...