2 papers
cs.CL2025
From KMMLU-Redux to KMMLU-Pro: A Professional Korean Benchmark Suite for LLM Evaluation
Seokhee Hong, Sunkyoung Kim, Guijin Son +3
The development of Large Language Models (LLMs) requires robust benchmarks that encompass not only academic domains but also industrial fields to effectively evaluate their applica…
cs.LG2025
Benchmarks Are Not Atomic: Composition-Aware LLM Evaluation using BenchHub
Eunsu Kim, Haneul Yoo, Guijin Son +3
LLM benchmarks are often treated as coherent measurement units, yet they are heterogeneous collections of instances spanning diverse domains, skills, formats, and contexts. As a re…