Publications

PolyMerge: Compressing 3D Gaussian Splats with Polytope Coverings for Provably Safe Resource-Constrained Navigation

[paper]

Published in IEEE Robotics and Automation Letters (RA-L); presented at IROS 2026, 2026

Authors: Jihoon Hong, C. Chiu, Sara Fridovich-Keil, Glen Chou

KLIP: Localized Distribution Shift Detection via KL-Divergence with Diffusion Priors in Inverse Problems

[paper]

Published in IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026

Authors: A. Kheirandish*, Jihoon Hong*, Sara Fridovich-Keil

LAMB: A Training-Free Method to Enhance the Long-Context Understanding of SSMs via Attention-Guided Token Filtering

[paper]

Published in Annual Meeting of the Association for Computational Linguistics (ACL), 2025

Authors: Zhifan Ye, Zheng Wang, Kejing Xia, Jihoon Hong, Leshu Li, Lexington Whalen, Cheng Wan, Yonggan Fu, Yingyan (Celine) Lin, Souvik Kundu

LongMamba: Enhancing Mamba’s Long-Context Capabilities via Training-Free Receptive Field Enlargement

[paper]

Published in International Conference on Learning Representations (ICLR), 2025

Authors: Zhifan Ye, Kejing Xia, Yonggan Fu, Xin Dong, Jihoon Hong, Xiangchi Yuan, Shizhe Diao, Jan Kautz, Pavlo Molchanov, Yingyan (Celine) Lin

3D Gaussian Rendering Can Be Sparser: Efficient Rendering via Learned Fragment Pruning

[paper]

Published in Conference on Neural Information Processing Systems (NeurIPS), 2024

Authors: Zhifan Ye, Chenxi Wan, Chaojian Li, Jihoon Hong, Sixu Li, Leshu Li, Yongan Zhang, Yingyan (Celine) Lin

VGA: Hardware Accelerator for Scalable Long Sequence Model Inference

[paper]

Published in IEEE/ACM International Symposium on Microarchitecture (MICRO), 2024

Authors: Seung Yul Lee, Hyunseung Lee, Jihoon Hong, SangLyul Cho, Jae W. Lee

A Memory-Efficient Edge Inference Accelerator with XOR-based Model Compression

[paper]

Published in Design Automation Conference (DAC), 2023

Authors: H. Lee, J. Hong, S. Kim, S. Y. Lee and J. W. Lee

Steering Robustness into World Action Models via Mechanistic Interpretability and Optimal Control

[paper]

Submitted to Conference on Robot Learning (CoRL), 2026

Authors: Jihoon Hong*, J. Skifstad*, Q. Dai, A. Chan, Glen Chou

Activation Steering of Video Generation Models via Reduced-Order Linear Optimal Control

[paper]

Submitted to Conference on Neural Information Processing Systems (NeurIPS), 2026

Authors: Jihoon Hong, A. Chan, Q. Dai, J. Skifstad, Glen Chou

LACToS: Local Anomaly Detection via Conditional Typicality of Superpixels

Submitted to Conference on Neural Information Processing Systems (NeurIPS), 2026

Authors: Jihoon Hong*, A. Kheirandish*, Sara Fridovich-Keil