Zhaohua Zhang — accepted AI-conference papers

2026

FailureAtlas: Mapping the Failure Landscape of T2I Models via Active Exploration

Muxi Chen, Zhaohua Zhang, Chenchen Zhao, Mingyang Chen, Wenyu Jiang, Tianwen Jiang +5

CVPR 2026

Static benchmark-driven evaluation has provided a valuable foundation for analyzing Text-to-Image (T2I) models.However, the fixed and predetermined prompt sets in benchmarks inherently limit diagnostic depth, making it difficult to uncover the full landscape of models' systematic failures or isolate

Cited by 0 Source ↗code ↗Scholar ↗

2026

GRAPE: Let GRPO Supervise Query Rewriting by Ranking for Retrieval

Zhaohua Zhang, Jianhuan Zhuo, Muxi Chen, Chenchen Zhao, Wenyu Jiang, Tianwen Jiang +5

ICML 2026poster

The CLIP model has established itself as a cornerstone of large-scale retrieval systems. However, its performance often degrades under distributional shifts such as multilingual, long-form, or multimodal queries. To avoid the prohibitive costs associated with retriever retraining or corpus re-embedd…

Cited by 0 Source ↗Scholar ↗