PrismML, a venture originating from Caltech, has introduced its new 1-bit large language model, Bonsai 8B, designed to significantly enhance AI efficiency on edge hardware. This innovative model architecture represents weights using only their sign and a shared scale factor, resulting in a memory footprint of just 1.15 GB. Compared to full-precision models, Bonsai 8B is 14 times smaller, 8 times faster, and 5 times more energy-efficient, while maintaining competitive performance. By drastically reducing memory and power requirements, PrismML aims to enable advanced AI applications on mobile devices, real-time robotics, and secure enterprise systems, effectively moving powerful language models out of massive cloud datacenters and onto local hardware.
Researchers have found that even seemingly random events, like the roll of a die, are governed by fundamental laws of physics. Their work provides further evidence for a long-held belief that the universe is fundamentally deterministic, even if it appears chaotic.
In essence, the study reinforces that the Boltzmann distribution isn't just *a* way to model randomness, it's *the* way to model truly independent random systems.
In essence, the study reinforces that the Boltzmann distribution isn't just *a* way to model randomness, it's *the* way to model truly independent random systems.
Caltech physicists have created the largest qubit array ever assembled: 6,100 neutral-atom qubits trapped in a grid by lasers. This milestone demonstrates a pathway to large error-corrected quantum computers, maintaining qubit superposition for about 13 seconds with 99.98% manipulation accuracy.
The SPHEREx mission will observe the entire sky multiple times over its planned two-year mission. It employs Linear-Variable Filter (LVF) technology to capture spectra across 100 spectral bands in the near-infrared spectrum. The mission focuses on exploring the origins of the universe, galaxies, water, and pre-biotic molecules, with a deep survey near the ecliptic poles and an all-sky survey during its mission. SPHEREx aims to achieve deeper sensitivities than 2MASS and meet scientific requirements with margins over instrument performance.
Matilde Marcolli, a mathematician at Caltech, has developed a new mathematical framework for Noam Chomsky's model of language, using ideas from theoretical physics. Chomsky's latest model, the minimalist program, simplifies language into a single computational operation called "merge," which combines elements of a sentence. Marcolli used Hopf algebras, a mathematical tool used in theoretical physics, to describe this merge operation. She worked with Chomsky and Bob Berwick to create a forthcoming book explaining this mathematical approach to language. Marcolli also taught a course and hosted workshops at Caltech on this topic, bringing together experts from various fields. This research supports Chomsky's philosophy that language should be studied using the methods and tools of the physical sciences.
