Publications

Song Park, Sanghyuk Chun, Byeongho Heo, Dongyoon Han (2025). DNNs May Determine Major Properties of Their Outputs Early, with Timing Possibly Driven by Bias. arXiv preprint arXiv:2502.08167.

Jaehui Hwang, Dongyoon Han, Byeongho Heo, Song Park, Sanghyuk Chun, Jong-Seok Lee (2024). Similarity of neural architectures using adversarial attack transferability. European Conference on Computer Vision.

Minhyun Lee, Song Park, Byeongho Heo, Dongyoon Han, Hyunjung Shim (2024). SeiT++: Masked Token Modeling Improves Storage-efficient Training. European Conference on Computer Vision.

Byeongho Heo, Song Park, Dongyoon Han, Sangdoo Yun (2024). Rotary position embedding for vision transformer. European Conference on Computer Vision.

Sanghyuk Chun, Wonjae Kim, Song Park, Sangdoo Yun (2024). Probabilistic Language-Image Pre-Training. arXiv preprint arXiv:2410.18857.

Song Park, Sanghyuk Chun, Byeongho Heo, Wonjae Kim, Sangdoo Yun (2023). SeiT: Storage-efficient vision training with tokens using 1% of pixel storage. Proceedings of the IEEE/CVF International Conference on Computer Vision.

Song Park, Sanghyuk Chun, Junbum Cha, Bado Lee, Hyunjung Shim (2022). Few-shot font generation with weakly supervised localized representations. IEEE transactions on pattern analysis and machine intelligence.

Sanghyuk Chun, Wonjae Kim, Song Park, Minsuk Chang, Seong Joon Oh (2022). Eccv caption: Correcting false negatives by collecting machine-and-human-verified image-caption associations for ms-coco. European Conference on Computer Vision.

Sanghyuk Chun, Song Park (2021). Styleaugment: Learning texture de-biased representations by style augmentation without pre-defined textures. arXiv preprint arXiv:2108.10549.

Song Park, Sanghyuk Chun, Junbum Cha, Bado Lee, Hyunjung Shim (2021). Multiple Heads are Better than One: Few-shot Font Generation with Multiple Localized Experts. International Conference on Computer Vision (ICCV).