MathScale.
Scaling instruction tuning for mathematical reasoning.
Large language models (LLMs) have demonstrated remarkable capabilities in problem-solving.
Continue reading “MathScale: Scaling Instruction Tuning for Mathematical Reasoning” »
Computer graphic simulations can represent natural phenomena such as tornados, underwater, vortices, and liquid foams more accurately thanks to an advancement in creating artificial intelligence (AI) neural networks.
Working with a multi-institutional team of researchers, Georgia Tech Assistant Professor Bo Zhu combined computer graphic simulations with machine learning models to create enhanced simulations of known phenomena. The new benchmark could lead to researchers constructing representations of other phenomena that have yet to be simulated.
Zhu co-authored the paper “Fluid Simulation on Neural Flow Maps.” The Association for Computing Machinery’s Special Interest Group in Computer Graphics and Interactive Technology (SIGGRAPH) gave it a best paper award in December at the SIGGRAPH Asia conference in Sydney, Australia.
Shortgpt layers in large language models are more redundant than you expect.
ShortGPT
Layers in large language models are more redundant than you expect.
We propose Strongly Supervised pre-training with ScreenShots (S4) — a novel pre-training paradigm for Vision-Language Models using data from large-scale web screenshot rendering.
Join the discussion on this paper page.
A pioneering Large Language Model for Law https://huggingface.co/papers/2403.
In this paper, we introduce SaulLM-7B, a large language model (LLM) tailored for the legal domain.
Join the discussion on this paper page.
GaLore.
Memory-efficient LLM training by gradient low-rank projection.
V/@animaanandkumar.
Continue reading “GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection” »
Apple presents How Far Are We from Intelligent Visual Deductive Reasoning?
Join the discussion on this paper page.
Researchers observe the quantum coherence of a quintet state with four electron spins in molecular systems for the first time at room temperature.
In a study published in Science Advances, a group of researchers led by Associate Professor Nobuhiro Yanai from Kyushu University’s Faculty of Engineering, in collaboration with Associate Professor Kiyoshi Miyata from Kyushu University and Professor Yasuhiro Kobori of Kobe University, reports that they have achieved quantum coherence at room temperature: the ability of a quantum system to maintain a well-defined state over time without getting affected by surrounding disturbances.
This breakthrough was made possible by embedding a chromophore, a dye molecule that absorbs light and emits color, in a metal-organic framework, or MOF, a nanoporous crystalline material composed of metal ions and organic ligands.
The last reported case of human plague in Florida occurred in 1920 during an outbreak in Pensacola.
Cases of leprosy have increased in Florida and the southeastern United States over the last decade, according to a new report.
Leprosy, officially called Hansen’s disease, is a rare type of bacterial infection that attacks the nerves and can cause swelling under the skin. The new research paper, published in the Centers for Disease Control and Prevention’s Emerging Infectious Diseases journal, found that reported cases doubled in the Southeast over the last 10 years.
Central Florida in particular has seen a disproportionate share of cases, which indicates it might be an endemic location for the disease, meaning leprosy has a consistent presence in the region’s population rather than popping up in the form of one-off outbreaks.
