2 papers
cs.CL2026
GRPO Beyond English: A Large-Scale Study of GRPO in Non-English and Multilingual Settings
Konstantin Dobler, Federico Scozzafava, Jonathan Janke +2
Reinforcement Learning with Verifiable Rewards (RLVR), often optimized with Group Relative Policy Optimization (GRPO), has become a central recipe for improving the reasoning capab…
cond-mat.supr-con2024
Maximum limit of connectivity in rectangular superconducting films with an oblique weak link
F. Colauto, D. Carmo, A. M. H. de Andrade +3
A method for measuring the electrical connectivity between parts of a rectangular superconductor was developed for weak links making an arbitrary angle with the long side of the sa…