About
Hi there! I’m Xijia Tao (Ciel, or 陶熙佳).
I am a PhD student at the HKUNLP lab. My research focuses on AI agents, particularly systems that can search for information, reason over multimodal evidence, and adapt to new tasks and environments.
More broadly, I’m interested in building increasingly capable AI systems and understanding how they can interact effectively with information, tools, and the world.
Publications
[arXiv] How Should Diffusion Language Models Edit Code? (2026)
Xijia Tao, Ziru Liu, Shansan Gong, Jiacheng Ye, Kecheng Chen, Zirui Wu, Lin Zheng, Xinyu Fu, Rui Liu, Lingpeng Kong
arXiv preprint arXiv:2609.38257
Paper
[arXiv] Visual Parallel Search: Learning to Search High-Resolution Images with Parallel Tile Inspection and Adaptive Zoom (2026)
Xijia Tao, Yihua Teng, Xinyu Fu, Cheng Gong, Ziru Liu, Xudong Xie, Rui Liu, Lingpeng Kong
arXiv preprint arXiv:2609.37002
Paper
[arXiv] SoftSkill: Behavioral Compression for Contextual Adaptation (2026)
Xijia Tao, Yihua Teng, Xinyu Fu, Ziru Liu, Kecheng Chen, Yuzhi Zhao, Suiyun Zhang, Rui Liu, Lingpeng Kong
arXiv preprint arXiv:2606.20333
Paper | Code
[arXiv] Self-Distilled Trajectory-Aware Boltzmann Modeling: Bridging the Training-Inference Discrepancy in Diffusion Language Models (2026)
Kecheng Chen, Ziru Liu, Xijia Tao, Hui Liu, Yibing Liu, Xinyu Fu, Shi Wu, Suiyun Zhang, Dandan Tu, Lingpeng Kong, Rui Liu, Haoliang Li
arXiv preprint arXiv:2605.11854
Paper
[ICML 2026] Beyond Confidence: Adaptive and Coherent Decoding for Diffusion Language Models (2026)
Kecheng Chen, Ziru Liu, Xijia Tao, Hui Liu, Xinyu Fu, Suiyun Zhang, Dandan Tu, Lingpeng Kong, Rui Liu, Haoliang Li
The Forty-third International Conference on Machine Learning (ICML 2026)
Project page | Paper | Proceedings
[ICLR 2026] MMSearch-Plus: Benchmarking Provenance-Aware Search for Multimodal Browsing Agents (2026)
Xijia Tao, Yihua Teng, Xinxing Su, Xinyu Fu, Jihao Wu, Chaofan Tao, Ziru Liu, Haoli Bai, Rui Liu, Lingpeng Kong
The Fourteenth International Conference on Learning Representations (ICLR 2026)
Project page | Paper | Code
[arXiv] OmniPlay: Benchmarking Omni-Modal Models on Omni-Modal Game Playing (2025)
Fuqing Bie, Shiyu Huang, Xijia Tao, Zhiqin Fang, Leyi Pan, Junzhe Chen, Min Ren, Liuyu Xiang, Zhaofeng He
arXiv preprint arXiv:2508.04361
Paper | Code
[NAACL 2025] ImgTrojan: Jailbreaking Vision-Language Models with ONE Image (2025)
Xijia Tao, Shuai Zhong, Lei Li, Qi Liu, Lingpeng Kong
Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL 2025)
Paper | Code
[arXiv] Language Versatilists vs. Specialists: An Empirical Revisiting on Multilingual Transfer Ability (2023)
Jiacheng Ye, Xijia Tao, Lingpeng Kong
arXiv preprint arXiv:2306.06688
Paper