Joonsik Nam

Integrated M.S./Ph.D. student at VDSLab, Sogang University

I am an integrated M.S./Ph.D. student at VDSLab, Sogang University, researching efficient visual AI systems. My work explores multimodal models and 3D/4D reconstruction, with an emphasis on making high-capacity vision systems faster, lighter, and practical beyond the lab. I am interested in carrying research ideas through to scalable implementations for real-world use.

  • Multi-modal LLM
  • 3D/4D reconstruction
  • Efficient vision
Download CV
Joonsik Nam

News

  1. TRiGS: Temporal Rigid-Body Motion for Scalable 4D Gaussian Splatting was accepted to the European Conference on Computer Vision (ECCV) 2026.

  2. OriCache: Orientation-Guided Feature Caching for DiT Acceleration was accepted to the International Conference on Machine Learning (ICML) 2026 AdaptFM Workshop.

Publications

* Equal contribution   † Corresponding author

International

ECCV 2026Latest

TRiGS: Temporal Rigid-Body Motion for Scalable 4D Gaussian Splatting

Suwoong Yeom*, Joonsik Nam*, Seunggyu Choi*, Lucas Yunkyu Lee, Sangmin Kim, Jaesik Park, Joonsoo Kim, Kugjin Yun, Kyeongbo Kong†, Suk-Ju Kang†

arXivProject page
Overview figure from OriCache: Orientation-Guided Feature Caching for DiT Acceleration

ICML 2026 Workshop

OriCache: Orientation-Guided Feature Caching for DiT Acceleration

Joonsik Nam, Hyunwoo Yu, Suk-Ju Kang†

OpenReview
Overview figure from Reducing Attention Distribution Error with Unified Tail Aggregation for Sparse Attention

ICML 2026 Workshop

Reducing Attention Distribution Error with Unified Tail Aggregation for Sparse Attention

Hyunwoo Yu*, Jongbeom Lee*, Jaemin Choi*, Jincheol Yang*, Yubin Cho*, Joonsik Nam, Seunghun Moon, Jung-Woo Chang, Bongjoon Hyun, Kyomin Sohn, Kyeongbo Kong†, Suk-Ju Kang†

OpenReview
Overview figure from HOIGS: Human-Object Interaction Gaussian Splatting

IEEE TMMUnder review

HOIGS: Human-Object Interaction Gaussian Splatting

Taewoo Kim*, Suwoong Yeom*, Jaehyun Pyun*, Geonho Cha, Dongyoon Wee, Joonsik Nam, Yun-Seong Jeong, Kyeongbo Kong†, Suk-Ju Kang†

arXiv

IEEE TETCIUnder review

SPIDFormer: Dual Low-Rank Attention For Spiking Neural Network

Hyunwoo Yu*, Joonsik Nam*, Jung-Woo Chang, Suk-Ju Kang†

Domestic

Overview figure from DiT 가속을 위한 방향 일관성 기반 특징 재사용 기법

대한전자공학회 2026 하계

DiT 가속을 위한 방향 일관성 기반 특징 재사용 기법

남준식, 유현우, 강석주†

Background

Education

  • Integrated M.S./Ph.D. Program

    Sep 2025 – Present

    Currently pursuing an integrated M.S./Ph.D. program at Sogang University.

  • Undergraduate

    Mar 2019 – Aug 2025

    Completed undergraduate studies in Electronic Engineering at Sogang University.

Projects

  • NAVER 공동 연구

    Jan 2026 – Present

    NAVER Cloud, Korea / 참여연구원

  • ETRI 아날로그-디지털 혼성 초저전력 뉴로모픽 엣지 SOC 개발

    Mar 2025 – Present

    ETRI, Korea / 참여연구원

Academic Service

  • Program Committee, AAAI

    2027

  • Reviewer, AAAI

    2026

Contact