Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
Josef Ressel Centre for Intelligent and Secure Industrial Automation, Salzburg University of Applied Sciences, Austria {stefan.huber,martin.uray}@fh-salzburg.ac.at
1
0
2
4
LLM judges inflate math proof scores by up to 0.36 points, revealing a significant alignment gap with human experts and a reasoning breakdown in discrete domains.