Web is increasingly becoming the primary platform to deliver AI services onto edge devices, making in-browser deep learning (DL) inference more prominent. Nevertheless, the heterogeneity of edge devices, combined with the underdeveloped state of Web hardware acceleration practices, hinders current in-browser inference from achieving its…
Temporal action abstractions, along with belief state representations, are a powerful knowledge sharing mechanism for sequential decision making. In this work, we propose a novel view that treats inducing temporal action abstractions as a sequence compression problem. To do so, we bring a subtle but…
Large Language Models (LLMs) are increasingly applied in various real-world scenarios due to their excellent generalization capabilities and robust generative abilities. However, they exhibit position bias, also known as"lost in the middle", a phenomenon that is especially pronounced in long-context scenarios, which indicates the placement…
Microsoft is using artificial intelligence to help preserve the history of D-Day, 80 years after allied forces landed on the beaches of Normandy, France, and turned the tide of World War II.
Aurora, a new AI foundation model from Microsoft Research, can transform our ability to predict and mitigate extreme weather events and the effects of climate change by enabling faster and more accurate weather forecasts than ever before.
Although diets influence health and the environment, measuring and changing nutrition is challenging. Traditional measurement methods face challenges, and designing and conducting behavior-changing interventions is conceptually and logistically complicated. Situated local communities such as university campuses offer unique opportunities to shape the nutritional environment and…
Training models with longer in-context lengths is a significant challenge for multimodal model due to substantial GPU memory and computational costs. This exploratory study does not present state-of-the-art models; rather, it introduces an innovative method designed to increase in-context text length in multi-modality large language…
Microsoft is honored to support the ACM Conference on Fairness, Accountability, and Transparency (ACM FAccT (opens in new tab)). This pivotal computer science conference has a multidisciplinary approach, uniting researchers and practitioners who are dedicated to advancing fairness, accountability, and transparency within socio-technical systems. Program…
Large language models (LLMs) have shown impressive capabilities across various tasks. However, training LLMs from scratch requires significant computational power and extensive memory capacity. Recent studies have explored low-rank structures on weights for efficient fine-tuning in terms of parameters and memory, either through low-rank adaptation…
Large language models (LLM) have achieved impressive performance on medical question-answering benchmarks. However, high benchmark accuracy does not imply that the performance generalizes to real-world clinical settings. Medical question-answering benchmarks rely on assumptions consistent with quantifying LLM performance but that may not hold in the…
In high-stakes domains like clinical reasoning, AI assistants powered by large language models (LLMs) are yet to be reliable and safe. We identify a key obstacle towards reliability: existing LLMs are trained to answer any question, even with incomplete context in the prompt or insufficient…
Robot learning requires a considerable amount of data to realize it's promise of generalization. However, it can be challenging to actually collect the magnitude of data necessary for generalization entirely in the real world. Simulation can serve as a source of plentiful data with coverage…
