Ziyu Guo

PhD student, Department of Computer Science and Engineering, The Chinese University of Hong Kong

6 papers at NeurIPS 2025

OpenReview· Semantic Scholar· Google Scholar

Poster Session 2

1 paper

Wednesday, December 3, 2025 · 4:30 PM → 7:30 PM

Exhibit Hall C,D,E

Delving into RL for Image Generation with CoT: A Study on DPO vs. GRPO

#5508 · Chengzhuo Tong, Ziyu Guo, Renrui Zhang, Wenyu Shan, Xinyu Wei, Zhenghao Xing, Hongsheng Li, Pheng-Ann Heng

We compare RL methods DPO & GRPO for image generation, showing their strengths & how rewards affect generalization. Explores scaling for better CoT-based synthesis.

Poster Session 3

3 papers

Thursday, December 4, 2025 · 11:00 AM → 2:00 PM

Exhibit Hall C,D,E

Perceive Anything: Recognize, Explain, Caption, and Segment Anything in Images and Videos

#4916 · Weifeng Lin, Xinyu Wei, Ruichuan An, Tianhe Ren, Tingwei Chen, Renrui Zhang, Ziyu Guo, Wentao Zhang, Lei Zhang, Hongsheng Li

UniCTokens: Boosting Personalized Understanding and Generation via Unified Concept Tokens

#5411 · Ruichuan An, Sihan Yang, Renrui Zhang, Zijun Shen, Ming Lu, Gaole Dai, Hao Liang, Ziyu Guo, Shilin Yan, Yulin Luo, Bocheng Zou, Chaoqun Yang, Wentao Zhang

What We Miss Matters: Learning from the Overlooked in Point Cloud Transformers

#5014 · Yi Wang, Jiaze Wang, Ziyu Guo, Renrui Zhang, Donghao Zhou, Guangyong Chen, Anfeng Liu, Pheng-Ann Heng

Poster Session 4

1 paper

Thursday, December 4, 2025 · 4:30 PM → 7:30 PM

Exhibit Hall C,D,E

T2I-R1: Reinforcing Image Generation with Collaborative Semantic-level and Token-level CoT

#5306 · Dongzhi Jiang, Ziyu Guo, Renrui Zhang, Zhuofan Zong, Hao Li, Le Zhuo, Shilin Yan, Pheng-Ann Heng, Hongsheng Li

Poster Session 6

1 paper

Friday, December 5, 2025 · 4:30 PM → 7:30 PM

Exhibit Hall C,D,E

Adaptive Classifier-Free Guidance via Dynamic Low-Confidence Masking

#3312 · Pengxiang Li, Shilin Yan, Jiayin Cai, Renrui Zhang, Ruichuan An, Ziyu Guo, Xiaowei Gao

We propose Adaptive Classifier-Free Guidance (A-CFG), which dynamically re-masks low-confidence tokens for a more targeted unconditional input in iterative models.