In the digital era where data is vast, the importance of information retrieval cannot be overstated, particularly for search engines, recommender systems, and applications that find documents based on their content. Information retrieval involves three fundamental challenges - relevance assessment, document ranking, and efficiency. BM25S is a recently introduced Python library that tackles these challenges…
Advancements in vision-language models (VLMs) have enabled the possibility of developing a fully autonomous Artificial Intelligence (AI) assistant that can perform daily computer tasks through natural language. However, just having the reasoning and common-sense abilities doesn't always lead to intelligent assistant behavior. Thus, a method to translate pre-training abilities into practical AI agents is crucial.…
Policymakers usually depend on coarse-resolution global climate models to assess a community's risk of extreme weather. By looking decades and even centuries into the future, these models can predict large-scale weather patterns but struggle to provide specific data for smaller locations. To estimate the risk of an area such as Boston experiencing extreme weather events…
Researchers from the OATML group at the University of Oxford have developed a statistical method to improve the reliability of large language models (LLMs) such as ChatGPT and Gemini. This method looks to mitigate the issues of "hallucinations," wherein the model generates false or unsupported information, and "confabulations," where the model provides arbitrary or incorrect…
Language Learning Models (LLMs) such as ChatGPT and Gemini have shown the capability of answering complex queries, but they often produce false or unsupported information, a situation aptly titled "hallucinations". This gets in the way of their reliability, with potential repercussions in critical fields like law and medicine. A specific subset of these hallucinations, known…
Neural networks, despite being theoretically capable of fitting as many data samples as they have parameters, often fall short in reality due to limitations in training procedures. This creates a gap between their potential and their practical performance, which can be an obstacle for applications that require precise data fitting, such as medical diagnoses, autonomous…
Researchers at MIT have developed a method that improves the accuracy of predictions generated by climate models. The technique involves the use of machine learning and dynamical systems theory to make predictions from coarse climate models more accurate. These models, which are used to predict the impact of climate change including extreme weather events, work…
The integration of artificial intelligence (AI) in clinical pathology represents an exciting frontier in healthcare, but key challenges include data constraints, model transparency, and interoperability. These issues prevent AI and machine learning (ML) algorithms from being widely adopted in clinical settings, despite their proven effectiveness in tasks such as cell segmentation, image classification, and prognosis…
Historically, thinking around decision-making has dichotomized habitual and goal-oriented behavior, treating them as independent activities controlled by distinct neural systems. Habitual behaviors, being automatic, are fast and model-free while goal-oriented behaviors, requiring deliberate action, are slower, model-based but demanding computationally. Microsoft researchers, however, have proposed an innovative Bayesian behavior framework that attempts to synergize these…
Scientists from MIT and the Pacific Northwest National Laboratory have developed a way to increase the accuracy of large-scale climate models, allowing for more precise predictions of extreme weather incidents in specific locations. Their process involves using machine learning in tandem with existing climate models to make the models' predictions closer to real-world observations. This…
Machine learning (ML) algorithms have increasingly found use in ecological modelling, including the prediction of Soil Organic Carbon (SOC), a critical component for soil health. However, their application in smaller datasets characteristic of long-term soil research still needs further exploration, notably in comparison with traditional process-based models. A study conducted in Austria compared the performance…
Artificial intelligence (AI) models today have become increasingly complex with billions of parameters. Existing AI models are largely inaccessible to many due to a lack of widespread knowledge of how to create and control them. MosaicML, a company co-founded by Jonathan Frankle PhD '23 and MIT Associate Professor Michael Carbin, strives to overcome this issue.…