Evaluates offline action prediction for Korean web agents across 549 tasks on 144 websites, pairing screenshots and action histories with human-recorded next-step targets.
Evaluates whether GUI agents can locate interface elements from Korean instructions and website screenshots. Its 3,319 human-annotated examples span 142 websites, providing a focused test of visual grounding on Korean web interfaces.
Evaluates whether web agents can complete tasks on live Korean websites. Its 312 human-written tasks span 104 sites, and success is judged by the final website outcome rather than a single prescribed interaction path.
Provides executable software-engineering repair tasks for evaluating coding agents. Its 2,955 synthetic instances span 53 Python repositories and include bug-injection patches, reference repairs, and test information.
Provides 5.6 million cleaned Uzbek documents—about 6.06 billion words across Latin and Cyrillic—for language-model pretraining and NLP. It combines books, archives, legal texts, web pages and public Telegram posts with source, topic, domain and register annotations.
Scales verifiable gameplay data by having a Builder and visual and coding players co-develop web games. Progress-weighted training improves agent performance on both its own benchmark and an external game benchmark.