Chaoda Song
PhD Student
Vision-Language-Action (VLA), Vision-Language Models (VLM), Large Language Models (LLM)
I am a PhD candidate in Computer Science and Data Science at Case Western Reserve University, advised by Prof. Yu Yin. My research focuses on the intersection of computer vision and large language models, with current work on VLA-driven robotic manipulation on real hardware. Outside of research, I enjoy tennis, music, and K-pop and hip-hop dance. I am always happy to connect, exchange ideas, and discuss new directions—feel free to reach out.
Education
- B.S. in Software Engineering, Zhengzhou University, China
- Ph.D. in Computer Science, Case Western Reserve University, USA
Research Interests
- Computer Vision
- Large Language Models (LLM)
- Vision-Language-Action (VLA) models
Academic Service
- Reviewer, ICML 2026
- Reviewer, ECCV 2026
- Reviewer, CoLM 2026
- Reviewer, ICME 2026
Publications
- Imagining Recovery: Inference-Time Counterfactual Realignment for Vision-Language-Action Models.In arXiv preprint arXiv:2608.14822, 2026.
@article{zhang2026imagining, title = {Imagining Recovery: Inference-Time Counterfactual Realignment for Vision-Language-Action Models}, author = {Zhang, Yanyan and Liu, Disheng and Ye, Kai and Song, Chaoda and Li, Xinpeng and Hariri, Mohsen and Singh, Vikash and Yin, Yu and Chaudhary, Vipin}, journal = {arXiv preprint arXiv:2608.14822}, year = {2026}, status = {preprint}, pdf = {https://arxiv.org/pdf/2608.14822.pdf} } - Overcoming Dynamics-Blindness: Training-Free Pace-and-Path Correction for VLA Models.In arXiv preprint arXiv:2605.11459, 2026.
@article{zhang2026dynamics, title = {Overcoming Dynamics-Blindness: Training-Free Pace-and-Path Correction for VLA Models}, author = {Zhang, Yanyan and Song, Chaoda and Singh, Vikash and Li, Xinpeng and Ye, Kai and Hu, Zhe and Pu, Zhongzhu and Yin, Yu and Chaudhary, Vipin}, journal = {arXiv preprint arXiv:2605.11459}, year = {2026}, status = {preprint}, pdf = {https://arxiv.org/pdf/2605.11459.pdf} } - Expo-GS: Exposure-Aware Signed Distance Function in Gaussian Splatting for High Dynamic Range.In Proceedings of the International Conference on Machine Learning (ICML), 2026.
@inproceedings{song2026expogs, title = {Expo-GS: Exposure-Aware Signed Distance Function in Gaussian Splatting for High Dynamic Range}, author = {Song, Chaoda and Lu, Yiren and Li, Xinpeng and Zhou, Yunlai and Zhang, Yanyan and Yin, Yu and Chaudhary, Vipin}, booktitle = {Proceedings of the International Conference on Machine Learning (ICML)}, year = {2026}, status = {accepted}, pdf = {https://openreview.net/forum?id=2itDye7kPO} } - CausalRAG2: Hierarchical Causal Knowledge Graph Design for RAG.In Proceedings of the International Conference on Machine Learning (ICML), 2026.
@article{wang2026causalrag, title = {CausalRAG2: Hierarchical Causal Knowledge Graph Design for RAG}, author = {Wang, Nengbo and Liang, Tuo and Singh, Vikash and Song, Chaoda and Yang, Van and Yin, Yu and Ma, Jing and Singh, Jagdip and Chaudhary, Vipin}, journal = {Proceedings of the International Conference on Machine Learning (ICML)}, year = {2026}, status = {accepted}, github = {https://github.com/Pwnb/CausalRAG2}, pdf = {https://arxiv.org/pdf/2602.05143.pdf} } - Segment then Splat: Unified 3D Open-Vocabulary Segmentation via Gaussian Splatting.In Advances in Neural Information Processing Systems (NeurIPS), 2025.
@article{lu2025segment, title = {Segment then Splat: Unified 3D Open-Vocabulary Segmentation via Gaussian Splatting}, author = {Lu, Yiren and Zhou, Yunlai and Qiao, Yiran and Song, Chaoda and Liang, Tuo and Ma, Jing and Wang, Huan and Yin, Yu}, journal = {Advances in Neural Information Processing Systems (NeurIPS)}, year = {2025}, status = {accepted}, pdf = {https://arxiv.org/pdf/2503.22204v2}, website = {https://yiren-lu.com/project_pages/Segment-then-Splat/}, code = {https://github.com/luyr/Segment-then-Splat} }