Appier Group, Inc.

Appier, an AI-native company listed on the Tokyo Stock Exchange, has published two research papers aimed at improving the reliability of enterprise Agentic AI by teaching large language models to recognize when retrieved information is insufficient and to select the appropriate reasoning language for each task. In one study, the team tested 28 leading LLMs and found that accuracy dropped by 30% to 50% when "none of the above" was the correct answer, but applying Direct Preference Optimization improved accuracy by nearly 30 percentage points. The other paper revealed that models often default to high-resource languages like English for reasoning, even when prompted in another language, and that local-language reasoning better captures cultural context and safety judgments. CEO Chih Han Yu emphasized that these capabilities will help Agentic AI evolve from executing instructions to making reliable autonomous decisions, and the research will be applied across Appier's Ad Cloud, Personalization Cloud, and Data Cloud product lines.