Publications

Selected Publications

Conference & Journal Papers

2026

• Jin Cui, Xinyue Long, Boran Zhao, et al. The Model Knows Another Way: Strategy Switching for Effective RLVR Exploration. https://doi.org/10.48550/arXiv.2609.33085, 2026.
• Jin Cui, Zhaoyu Pu, Botao Cai, et al. StrataVLA: Hierarchical and Efficient 3D Geometric Grounding for Vision-Language-Action Models. https://doi.org/10.48550/arXiv.2609.26071, 2026.
• Jin Cui, Chuanchang Su, Jiayi Lu, et al. HAFI-VLM: A Frequency Perspective for Diagnosing and Enhancing Visual Perception in Vision-Language Models. https://doi.org/10.48550/arXiv.2608.02124, 2026.
• Jin Cui, Yanbin Hu, Xinyue Long, et al. Look Where It Matters: Adaptive Visual Refinement for Vision-Language-Action Models. https://doi.org/10.48550/arXiv.2608.02197, 2026.
• Yanbin Hu, Jin Cui, Jun Ye, et al. Distill What RGB Can Recover: Privileged 3D Evidence for RGB-Only Vision-Language Models. https://doi.org/10.48550/arXiv.2608.00110, 2026.
• Hetian Liu, Jin Cui, Mengcheng Shi, et al. GLOBE: Trajectory-Aligned Gradient Matching with Structured SparseOptimization for Coreset Selection. https://doi.org/10.48550/arXiv.2608.02690, 2026.
• Yanbin Hu, Jin Cui, Jiayi Lu, et al. ECHO: Continuous Hierarchical Memory for Vision-Language-Action Models. Annual Conference on Neural Information Processing Systems (NeurIPS) , 2026.
• Boran Zhao, Hetian Liu, Zhenxian Hu, et al. CAST: Collapse-Aware multi-Scale Topology Fusion for Multimodal Coreset Selection. https://doi.org/10.48550/arXiv.2605.11705, 2026.
• Jin Cui, Xinyue Long, Xunyong Zhang, et al. Retrieve, Integrate, and Synthesize: Spatial-Semantic Grounded Latent Visual Reasoning. https://doi.org/10.48550/arXiv.2605.07106, 2026.
• Jin Cui, Jiaqi Guo, Jiepeng Zhou, et al. "The Whole Is Greater Than the Sum of Its Parts": A Compatibility-Aware Multi-Teacher CoT Distillation Framework. 35th International Joint Conference on Artificial Intelligence (IJCAI), 2026.
• Jin Cui, Jiaqi Guo, Jiepeng Zhou, et al. MIND: From Passive Mimicry to Active Reasoning through Capability-Aware Multi-Perspective CoT Distillation. The 64th Annual Meeting of the Association for Computational Linguistics (ACL)., 2026.
• Jin Cui, Boran Zhao, Jiajun Xu, Jiaqi Guo, et al. FAST: Topology-Aware Frequency-Domain Distribution Matching for Coreset Selection. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , 2026.

2025

• Boran Zhao et al. SparseMap: A Sparse Tensor Accelerator Framework Based on Evolution Strategy. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD), 2025.
• Boran Zhao, Zihang Yuan, Yanbin Hu, et al. IsoSched: Preemptive Tile Cascaded Scheduling of Multi-DNN via Subgraph Isomorphism. IEEE Transactions on Computer-Aided Design of Integrated Circuit and Systems (TCAD), 2025.
• Boran Zhao, Hetian Liu, Zihang Yuan, et al. AdapSNE: Adaptive Fireworks-Optimized and Entropy-Guided Dataset Sampling for Edge DNN Training. IEEE Transactions on Circuits and Systems I: Regular Papers (TCAS-1), 2025.

2024

Book Chapters

Newsletters