VT-Bench: Benchmark for Visual-Tabular Multi-Modal AI
Discover VT-Bench, the first unified benchmark for visual-tabular multi-modal learning with 14 datasets across 9 domains, advancing AI research.
Boosting Vision Language Models with Self-Captioning Tuning
Discover how self-captioning multimodal tuning amplifies redundancies to enhance robustness and reduce errors in vision language models.
NoiseRater: Enhancing Diffusion Model Training with Noise Valuation
Discover NoiseRater, a meta-learning framework that improves diffusion model training by valuing noise samples for faster, higher-quality results.
HoReN: Scalable Model Editing for Large Language Models
Discover HoReN, a scalable normalized Hopfield retrieval method for efficient, accurate large-scale sequential model editing in AI systems.
Deep Learning Forecasts Stability in Tritium Experiments
Explore how deep learning models improve forecasting of source stability in tritium experiments, enhancing precision in scientific research.