VAIV x Korea Association for Information Science Next Generation Artificial Intelligence Contest Grand Prize Winner / Prof. Lee Jae Koo(Department of Artificial Intelligence) and his research team
- 24.07.23 / 박서연
Song Seung Heon, Seo Jung Hyun, and Kim Min Seong of the Master of Computer Science program in the Graduate School of General Studies, Department of Electronic Engineering (Artificial Intelligence Laboratory, advisor Lee Jae Koo) won the grand prize and received 2 million won at the VAIV x Korea Association of Information Science and Technology Generated AI Contest held at the 2024 Korea Computer Science Conference.
The VAIV x Korea Information Science Association Generative AI Contest was held to improve the performance of Korean search summarization using LLM. The AI Lab team merged the Knowledge Distillation method and Preference Alignment method using Oracle models. This method effectively generates high-quality data while allowing the model to answer in a way that users prefer. The method was well received as it effectively improved performance without human intervention.
This content is translated from Korean to English using the AI translation service DeepL and may contain translation errors such as jargon/pronouns. If you find any, please send your feedback to kookminpr@kookmin.ac.kr so we can correct them.
|
VAIV x Korea Association for Information Science Next Generation Artificial Intelligence Contest Grand Prize Winner / Prof. Lee Jae Koo(Department of Artificial Intelligence) and his research team |
||||
---|---|---|---|---|
2024-07-23
2267
Song Seung Heon, Seo Jung Hyun, and Kim Min Seong of the Master of Computer Science program in the Graduate School of General Studies, Department of Electronic Engineering (Artificial Intelligence Laboratory, advisor Lee Jae Koo) won the grand prize and received 2 million won at the VAIV x Korea Association of Information Science and Technology Generated AI Contest held at the 2024 Korea Computer Science Conference.
The VAIV x Korea Information Science Association Generative AI Contest was held to improve the performance of Korean search summarization using LLM. The AI Lab team merged the Knowledge Distillation method and Preference Alignment method using Oracle models. This method effectively generates high-quality data while allowing the model to answer in a way that users prefer. The method was well received as it effectively improved performance without human intervention.
|