Like human brains, large language models reason about diverse data in a general way
A new study shows LLMs represent different data types based on their underlying meaning and reason about data in their dominant language.
Language
LLMs develop their own understanding of reality as their language abilities improve
In controlled experiments, MIT CSAIL researchers discover simulations of reality developing deep within LLMs, indicating an understanding of language beyond simple mimicry.
MIT researchers advance automated interpretability in AI models
MAIA is a multimodal agent that can iteratively design experiments to better understand various components of AI systems.
Reasoning skills of large language models are often overestimated
New CSAIL research highlights how LLMs excel in familiar scenarios but struggle in novel ones, questioning their true reasoning abilities versus reliance on memorization.
Helping nonexperts build advanced generative AI models
MosaicML, co-founded by an MIT alumnus and a professor, made deep-learning models faster and more efficient. Its acquisition by Databricks broadened that mission.
Technique improves the reasoning capabilities of large language models
Combining natural language and programming, the method enables LLMs to solve numerical, analytical, and language-based tasks transparently.
AI agents help explain other AI systems
MIT researchers introduce a method that uses artificial intelligence to automate the explanation of complex neural networks.
Complex, unfamiliar sentences make the brain’s language network work harder
A new study finds that language regions in the left hemisphere light up when reading uncommon sentences, while straightforward sentences elicit little response.
Multi-AI collaboration helps reasoning and factual accuracy in large language models
Researchers use multiple AI models to collaborate, debate, and improve their reasoning abilities to advance the performance of LLMs while increasing accountability and factual accuracy.