Minsoo Kim
minsoo2333 [at] hanyang.ac.kr
![mskim_230127_idcard.jpg](/assets/img/mskim_230127_idcard.jpg)
Hi! I am Minsoo Kim, Ph.D. student in the AI Hardware & Algorithm lab at Hanyang University advided by professor Jungwook Choi and working as a PhD intern in Qualcomm AI Research.
My core research centers on enhancing the efficient algorithm for generative models, with a specific focus on understanding the effects of Quantization and Distillation across diverse NLP tasks aimed for efficient inference of LLMs. Beyond, I am passionated into advanced techniques for fine-tuning LLMs, aiming to fully unlock the potential of specialized LLMs.
News
May 2024 | 2 papers (1 main and 1 findings) accepted @ ACL 24 🇹🇭 |
---|---|
Mar 2024 | I am starting as a PhD Intern in Qualcomm AI Research ![]() |
Nov 2023 | Selcted as a winner in Qualcomm Innovation Fellowship 2023 ![]() |
Oct 2023 | 1 paper accepted and attending @ EMNLP 23 🇸🇬 1 paper accepted and attending @ NeurIPS 23 🇺🇸 |
Selected Publications
- ACL 2024RA-LoRA: Rank-Adaptive Parameter-Efficient Fine-Tuning for Accurate 2-bit Quantized Large Language ModelsIn Findings of the Association for Computational Linguistics: ACL 2024, Aug 2024
- ACL 2024Improving Conversational Abilities of Quantized Large Language Models via Direct Preference AlignmentIn Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics, Aug 2024