Nam Cao

M.S. Student in AI

KAIST AI · OSI Lab

I am Nam Cao, a Master’s student in KAIST’s Graduate School of AI, advised by Prof. Se-Young Yun in the OSI Lab. I earned my B.S. in Computer Science at KAIST.

My research centers on efficient LLM inference, including speculative decoding and block-diffusion drafting, as well as adversarial attacks on large vision-language models.

Recently, I’ve been especially interested in diffusion language models, which enable ultra-fast, parallel generation and challenge the traditional autoregressive paradigm, along with RL post-training methods to train such models efficiently.

Feel free to email me if you’d like to chat.

News

May 2026 🎉 My first paper as co-first author, BASTION, was selected for an oral presentation at the ICML 2026 AdaptFM workshop (Resource-Adaptive Foundation Model Inference)!
Feb 2026 🎉 We released our paper SAGA, on attention-guided adversarial attacks against vision-language models, on arXiv.
2025 🏆 Awarded the KAIST Global Presidential Scholarship (KGPS).
Feb 2025 🏫 Started my M.S. at KAIST AI, joining OSI Lab under Prof. Se-Young Yun.
Aug 2024 🎓 Graduated from KAIST with a B.S. in Computer Science.

Education

M.S. in Graduate School of AI, KAIST. Advised by Prof. Se-Young Yun Feb 2025 – Present
B.S. in Computer Science, KAIST. Aug 2020 – Aug 2024

Publications

^* equal contribution

2026

BASTION: Budget-Aware Speculative Decoding with Tree-structured Block Diffusion Drafting

Nam Cao^*, Soowon Oh^*, Yujin Kim, Hojung Jung, Huzama Ahmad, Sangmin Bae, Se-Young Yun

Resource-Adaptive Foundation Model Inference (AdaptFM) Workshop, ICML 2026 Oral

arXiv Code
Stage-wise Attention-Guided Region Sequencing for Adversarial Attacks on Large Vision-Language Models (SAGA)

Jaehyun Kwak, Nam Cao, Boryeong Cho, Segyu Lee, Sumyeong Ahn, Se-Young Yun

Manuscript under review, 2026

arXiv

Research Projects

On-Device Speculative Decoding 2025 – Present

Samsung Research (SoC Architecture team) × OSI Lab, KAIST — Co-Project Manager

Building an evaluation framework and speculative-decoding strategies for faster on-device LLM inference.
Deep Learning for Air-Quality Forecasting 2025 – Mar 2026

National Institute of Environmental Research (NIER), Korea × OSI Lab, KAIST

Training deep-learning models that forecast particulate-matter (PM) levels.

Awards & Honors

KAIST Global Presidential Scholarship (KGPS) — Four-semester graduate scholarship with a monthly living stipend. (2025–2027)
KAIST–Qualcomm Innovation Award — Winning team at the KAIST–Qualcomm AI Hackathon. (2023)
KAIST International Undergraduate Scholarship — Full-ride scholarship. (2020)
Vietnam Mathematical Olympiad (VMO) — Bronze Medal. (2018, 2019)