4+ hour, 2+ min ago   (403+ words) MIT researchers have identified significant examples of machine-learning model failure when those models are applied to data other than what they were trained on, raising questions about the need to test whenever a model is deployed in a new setting....

1+ mon, 2+ day ago   (475+ words) MIT CSAIL study suggests that neural network architectures considered unsuitable for modern tasks can improve with short-term guidance. The method encourages a target network to match a guide network's internal representations, improving its starting point and making machine learning easier....

1+ mon, 2+ day ago   (601+ words) Large language models struggle with state changes that are common in long texts, like how a cat might interact with a box over time and how the box might break down. Now, work from MIT-IBM Watson AI Lab researchers can…...

1+ mon, 3+ day ago   (751+ words) Why did humans evolve the eyes we have today? While scientists can't go back in time to study the environmental pressures that shaped the evolution of the diverse vision systems that exist in nature, a new computational framework developed by…...

1+ mon, 4+ day ago   (982+ words) Computer-aided design (CAD) systems are tried-and-true tools used to design many of the physical objects we use each day. But CAD software requires extensive expertise to master, and many tools incorporate such a high level of detail they don't lend…...

1+ mon, 5+ day ago   (231+ words) Q: What drew you to research microbes in extreme environments, and what are the challenges in studying them?" Q: Given how diverse microbes are and how little we understand about them, how can studying microbes in silico, using genomic language modeling, advance…...

1+ mon, 1+ week ago   (651+ words) A new approach developed by MIT CSAIL researchers uses an LLM to plan how to answer complex reasoning tasks, then divides the legwork of that strategy among smaller language models. Their method helps LMs provide more accurate responses than leading…...

1+ mon, 1+ week ago   (773+ words) Let's say an environmental scientist is studying whether exposure to air pollution is associated with lower birth weights in a particular county. They might train a machine-learning model to estimate the magnitude of this association, since machine-learning methods are especially…...

1+ mon, 1+ week ago   (310+ words) The School of Science welcomed 11 new faculty members in 2024. Fan earned her PhD at Harvard University after undergraduate studies at Peking University in China. She joined the MIT Department of Brain and Cognitive Sciences as the Samuel A. Goldblith Career Development…...

1+ mon, 2+ week ago   (710+ words) To make large language models (LLMs) more accurate when answering harder questions, researchers can let the model spend more time thinking about potential solutions. But common'approaches that give LLMs this capability set a fixed computational budget for every problem, regardless…...