About
Research & Development
Precision 3D Editing: New Framework Enables Detailed Part-Level Modifications in Gaussian Splatting
TLDR: RoMaP is a new 3D editing framework that allows precise and drastic modifications to specific parts of 3D Gaussian Splatting scenes. It achieves...
Research & Development
Optimizing Reference Signal Allocation in Massive MIMO Systems Through AI-Driven Channel Prediction
TLDR: This research introduces Channel Prediction-Based Reference Signal Allocation (CPRS), a novel deep learning approach that uses a ViViT/CNN model to predict optimal DM-RS...
Research & Development
LogTinyLLM: Making Log Anomaly Detection Efficient with Small Language Models
TLDR: LogTinyLLM is a new framework that uses parameter-efficient fine-tuning (PEFT) techniques, specifically LoRA and adapter methods, with 'tiny' large language models (LLMs) to...
Research & Development
EDITGEN: Revolutionizing Audio Editing with Instruction-Based Auto-Regressive Models
TLDR: EDITGEN is a novel audio editing framework that leverages cross-attention control within auto-regressive models, inspired by image editing techniques. It introduces three mechanisms...
Research & Development
GRNGC: A Unified Model for Discovering Causal Relationships in Complex Data
TLDR: GRNGC (Gradient Regularization-based Neural Granger Causality) is a new method for identifying causal relationships in time series data. Unlike previous models that require...
Research & Development
SystolicAttention: A Breakthrough in AI Accelerator Design
TLDR: A new hardware architecture called FSA, combined with the SystolicAttention algorithm, allows the entire FlashAttention process in AI models to run efficiently on...