About
Research & Development
Bridging Perception and Action: GF-VLA Enables Smart Dual-Arm Robot Control
TLDR: GF-VLA is a new robotics framework that teaches dual-arm robots complex manipulation skills from human video demonstrations. It uses information theory to understand...
Research & Development
ECHO System Decouples RL Training and Inference for Large Language Models
TLDR: ECHO is a novel Reinforcement Learning (RL) system that separates the trajectory sampling (inference) and policy optimization (training) phases for large language models...
Research & Development
QCopilot: An AI Framework for Quantum Experiment Automation
TLDR: QCopilot is an AI-powered multi-agent framework that uses large language models to automate the design, optimization, and diagnosis of quantum sensors. It integrates...
Research & Development
Unveiling Cultural Understanding in AI: The MyCulture Benchmark
TLDR: The MyCulture benchmark is introduced to evaluate Large Language Models (LLMs) on their understanding of Malaysia's diverse culture, presented in Bahasa Melayu. It...
Research & Development
Understanding Vision-Language Model Decisions Through Interaction Explanations
TLDR: A new research method, FIXLIP, uses game theory and weighted Banzhaf interactions to explain how vision-language models like CLIP determine image-text similarity. Unlike...
Research & Development
Measuring AI Confidence: How Sampling Frequency Improves Reliability in Multiple-Choice Questions
TLDR: This research introduces a frequency-based Predictive Entropy method combined with Conformal Prediction to quantify uncertainty in Large Language Models (LLMs) for multiple-choice question...