Sunchan Park

Postdoctoral Researcher | Pusan National University

sunchanpark@pusan.ac.kr | Google Scholar | ORCID | LinkedIn

Sunchan Park

I am a Postdoctoral Researcher in the Computer Vision & Signal Processing Lab at Pusan National University, working with Prof. Kyeongbo Kong. My background is in speech processing — speech, speaker, and emotion recognition. Building on this, I am expanding into multimodal AI, bridging audio, vision, and language, with a recent focus on spatio-temporal audio-visual understanding.

My research interests span Multimodal AI, Audio-Visual Perception, Spatio-Temporal Reasoning, and Agentic Systems.

Publications

NeurIPS-W

OmniWhere: Agentic Audio-Visual Perception for Embodied Spatial Reasoning

Sunchan Park and Kyeongbo Kong

In NeurIPS 2026 Workshop on Embodied Spatial Reasoning, 2026

Bib

@inproceedings{park2026omniwhere,
  title = {OmniWhere: Agentic Audio-Visual Perception for Embodied Spatial Reasoning},
  author = {Park, Sunchan and Kong, Kyeongbo},
  booktitle = {NeurIPS 2026 Workshop on Embodied Spatial Reasoning},
  year = {2026},
}

ICML-W

Compositional Agentic Formulation Search for Open-Vocabulary Audio-Visual Event Localization

Beomgwon Jo^*, Sunchan Park^*, and Kyeongbo Kong

In ICML 2026 Workshop on Compositional Learning, 2026

HTML PDF Bib

@inproceedings{jo2026compositional,
  title = {Compositional Agentic Formulation Search for Open-Vocabulary Audio-Visual Event Localization},
  author = {Jo, Beomgwon and Park, Sunchan and Kong, Kyeongbo},
  booktitle = {ICML 2026 Workshop on Compositional Learning},
  year = {2026},
}

SP Letters

Group-wise Layer Aggregation of Self-Supervised Representations for Speech Emotion Recognition

Sunchan Park, Hyung Soon Kim, and Kyeongbo Kong

IEEE Signal Processing Letters, 2026

HTML Bib

@article{park2026groupwise,
  title = {Group-wise Layer Aggregation of Self-Supervised Representations for Speech Emotion Recognition},
  author = {Park, Sunchan and Kim, Hyung Soon and Kong, Kyeongbo},
  journal = {IEEE Signal Processing Letters},
  volume = {33},
  pages = {2705--2709},
  year = {2026},
  doi = {10.1109/LSP.2026.3707494},
}

IEEE Access

Dimension-Specific Margins and Element-Wise Gradient Scaling for Enhanced Matryoshka Speaker Embedding

Sunchan Park and Hyung Soon Kim

IEEE Access, 2025

HTML Bib

@article{park2025dimension,
  title = {Dimension-Specific Margins and Element-Wise Gradient Scaling for Enhanced Matryoshka Speaker Embedding},
  author = {Park, Sunchan and Kim, Hyung Soon},
  journal = {IEEE Access},
  volume = {13},
  pages = {45473--45487},
  year = {2025},
}

El. Letters

Reverberant Environment Embedding using Dereverberation Autoencoder

Sunchan Park and Hyung Soon Kim

Electronics Letters, 2025

HTML Bib

@article{park2025reverberant,
  title = {Reverberant Environment Embedding using Dereverberation Autoencoder},
  author = {Park, Sunchan and Kim, Hyung Soon},
  journal = {Electronics Letters},
  volume = {61},
  number = {1},
  pages = {e70134},
  year = {2025},
  doi = {10.1049/ell2.70134},
}

Preprints

arXiv

CueRator: Agentic Search for Symbolic Rules to Adapt Frozen Multimodal Encoders

Sunchan Park^*, Beomkwon Cho^*, and Kyeongbo Kong

arXiv preprint arXiv:2610.07868, 2026

HTML PDF Code Bib

@article{park2026cuerator,
  title = {CueRator: Agentic Search for Symbolic Rules to Adapt Frozen Multimodal Encoders},
  author = {Park, Sunchan and Cho, Beomkwon and Kong, Kyeongbo},
  journal = {arXiv preprint arXiv:2610.07868},
  year = {2026},
}

Education

Ph.D. in Electronics Engineering, Pusan National University · 2019–2025
M.S. in Electronics Engineering, Pusan National University · 2016–2018
B.S. in Electronics Engineering, Pusan National University · 2010–2016

Experience

Postdoctoral Researcher, Pusan National University · 2025–present
Research Associate, LG Electronics · 2018–2019