2 papers
cs.CL2025
Decoding-Free Sampling Strategies for LLM Marginalization
David Pohl, Marco Cognetta, Junyoung Lee +1
Modern language models operate on subword-tokenized text in order to make a trade-off between model size, inference speed, and vocabulary coverage. A side effect of this is that, d…
physics.ins-det2019
Compact, directional neutron detectors capable of high-resolution nuclear recoil imaging
I. Jaegle, P. M. Lewis, M. Garcia-Sciveres +10
We report on the design, production, and performance of compact 40-cm Time Projection Chambers (TPCs) that detect fast neutrons by measuring the three-dimensional (3D) ionizati…