Two AI Papers by Elice Accepted to ACL 2025 Main Conference, World’s Premier NLP Conference

Two papers on AI models for education to be presented at the main conference, held from July 28 to 30

AI education solution company Elice, Inc. has received international recognition for the innovation of its education-specialized AI model research, with two papers accepted to ACL (Association for Computational Linguistics) 2025, the world’s most prestigious conference in the field of natural language processing (NLP).

Two AI research papers authored by Elice AI engineers Doyoun Kim and Yooseop Lee have been selected for presentation at the ACL 2025 main conference. This year’s ACL conference will be held in Vienna, Austria, from July 27 to August 1, with the main conference sessions taking place July 28–30.

Proposals include a deep learning model for more accurate prediction of student learning levels, and an LLM-based method for generating highly discriminative multiple-choice questions

The selected research focuses on enhancing the effectiveness of education through AI technologies.

• Doyoun Kim’s paper, titled “Knowledge Tracing in Programming Education Integrating Student’s Questions,” proposes a deep learning model that analyzes students’ problem-solving processes by incorporating question data asked during learning. This allows for faster and more precise prediction of individual learning levels, paving the way for personalized instruction.

  
  

• Yooseop Lee’s paper, “Generating Plausible Distractors for Multiple-Choice Questions via Student Choice Prediction,” introduces a method that augments large language models (LLMs) trained on student answer data with a pipeline for generating plausible incorrect options. This enhances the discriminative power of multiple-choice questions, enabling a more accurate assessment of learning.

  
  

AI Engineer Doyoun Kim commented, “I believe this research goes beyond mere technical improvement of AI models—it contributes to building an environment where all learners can ask questions, be understood, and grow in their own way.” AI Engineer Yooseop Lee added, “We’ll continue conducting AI research that provides real benefits to both learners and educators in actual educational settings.”

Suin Kim, Chief Research Officer (CRO) at Elice, Inc. and a co-author on both papers, remarked: “It’s a significant achievement for Elice to have our AI modeling capabilities, developed with a focus on the education sector, recognized at ACL 2025, one of the most globally respected conferences.” He added, “We will continue leading the way in AI model research and development, aiming to bring technological innovation to the educational environment and realize our vision of ‘a classroom for all.’”