KRETA, a 2,577-item Korean text-rich VQA benchmark, shows vision-language models recognize Korean text well but lag in multi-step reasoning, especially in open-source models.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
KRETA: A Benchmark for Korean Reading and Reasoning in Text-Rich VQA Attuned to Diverse Visual Contexts
KRETA, a 2,577-item Korean text-rich VQA benchmark, shows vision-language models recognize Korean text well but lag in multi-step reasoning, especially in open-source models.