ML identifies discrepancies in 40 years of nuclear data
bravo_abad · x · 2026-08-28
Researchers used a sparse Bayesian model to analyze 21 measurements of the prompt neutron spectrum from californium-252 fission spanning four decades. By encoding 41 experimental details, the ML identified features most associated with discrepancies. Physicists then used these clues to run simulations and new experiments, correcting or rejecting historical data and significantly reducing the spread in measurements.
More from Research
- Viewpoint: Optimizing for language compression led to current AI level — kuchaev · 2026-08-29
- New theory: Aging begins when cells lose shared developmental goals — rand_longevity · 2026-08-29
- SIGCOMM 2026 data: China now produces 57% of papers, surpassing US — vishalmisra · 2026-08-29
- MIT Prof. Patrick Winston Breaks Down the Simplest Neural Network — MIT_CSAIL · 2026-08-29
- First Text-to-SQL Model Beats Human Benchmark Using RLVR — EchoShao8899 · 2026-08-28
- Paper: Sol 5.6 Achieves 46% Accuracy on BrowseComp-Plus Without Search — beirmug · 2026-08-28