Stanford researchers have improved on current LLM optimization methods called Sophia that cuts the pretraining time in half. The team hopes to see Sophia applied to other areas of machine learning such as computer vision models or multi-modal models
This report documents the progress that is being made in expanding the extraction and processing of rare earth elements and lithium. While progress has been made in extraction and processing in Australia the United States and elsewhere, there is still much work to be done to assure a sufficient and reliable supply of critical materials.
Improved coordination of demands and resources on the electric grid could be achieved by installing software in homes and business. Stanford researchers say that coordination, if implemented widely, can accomplish more at far less expense and costly infrastructure upgrades
Stanford researchers evaluate whether major foundation model providers comply with the proposed EU law on AI. They provide recommendations that policymakers prioritize transparency and providers take action to collectively set industry standards that improve transparency.
A recent paper by Stanford researchers, Foundation Models and Fair Use, provides a historical context of fair use and lays out several hypotheticals to illustrate the knotty issues AI raises.
SLAC scientist Johanna Nelson Weker and her team are focused on finding ways to design batteries in a sustainable way, such as by using abundant materials rather than lithium and cobalt, which are more limited. In this Q&A, Nelson Weker tells us what the future of battery design could look like.
The Stanford Energy Research Year in Review, provides a snapshot of key studies from scholars across Stanford’s seven schools. These publications demonstrate how Stanford faculty, students, postdoctoral scholars, and research staff are building connections between knowledge generation and scalable impact.
Join The Future of Everything host Russ Altman in this episode with Matteo Cargnello, assistant professor in chemical engineering, who is working to turn harmful chemicals in the skies into other benign or useful chemicals.
A Stanford Graduate School of Business course looks at the importance of older consumers and workers and challenging assumptions based on outmoded ideas about lifespan and "health span". BMW and Merrill Lynch have redesigned their offerings to target this shift in customers who account for half of consumer spending worldwide.
Stanford engineers have found a metallic compound that could bring more efficient forms of computer memory closer to commercialization which facilitate a form of working memory that stores data in electron spin directions. This will enable faster processing and allow AI training to happen on individual devices instead of remote servers
