DynT2I-Eval: A Dynamic Evaluation Framework for Text-to-Image Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917468064186368 |
|---|---|
| author | Wang, Juntong Wang, Jiarui Duan, Huiyu Li, Lewei Zhai, Guangtao Min, Xiongkuo |
| author_facet | Wang, Juntong Wang, Jiarui Duan, Huiyu Li, Lewei Zhai, Guangtao Min, Xiongkuo |
| contents | Existing text-to-image (T2I) benchmarks largely rely on fixed prompt sets, leaving them vulnerable to overfitting and benchmark contamination once publicly released and repeatedly reused. In this work, we propose DynT2I-Eval, a fully automated dynamic evaluation framework for T2I models. It constructs a structured visual semantic space from long-form descriptions, decomposing prompts into controllable dimensions (e.g., subject, logical constraint, environment, and composition). This enables the continuous generation of fresh prompts via task-specific spaces and difficulty-aware sampling. DynT2I-Eval evaluates model performance across text alignment, perceptual quality, and aesthetics. Heterogeneous outputs are unified into prompt-conditioned pairwise comparisons, allowing a dynamic scheduler, micro-batch aggregation, and weighted Bayesian updates to maintain a stable online leaderboard despite changing prompt distributions and model injection. Experiments with independently sampled prompt streams demonstrate that continually refreshed prompts provide a robust evaluation protocol, reducing the impact of prompt-set-specific tuning. Simulations and ablations further confirm that the proposed ranking framework achieves a strong balance among cold-start convergence, late-entry discovery, and long-run ranking fidelity. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_06170 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | DynT2I-Eval: A Dynamic Evaluation Framework for Text-to-Image Models Wang, Juntong Wang, Jiarui Duan, Huiyu Li, Lewei Zhai, Guangtao Min, Xiongkuo Computer Vision and Pattern Recognition Existing text-to-image (T2I) benchmarks largely rely on fixed prompt sets, leaving them vulnerable to overfitting and benchmark contamination once publicly released and repeatedly reused. In this work, we propose DynT2I-Eval, a fully automated dynamic evaluation framework for T2I models. It constructs a structured visual semantic space from long-form descriptions, decomposing prompts into controllable dimensions (e.g., subject, logical constraint, environment, and composition). This enables the continuous generation of fresh prompts via task-specific spaces and difficulty-aware sampling. DynT2I-Eval evaluates model performance across text alignment, perceptual quality, and aesthetics. Heterogeneous outputs are unified into prompt-conditioned pairwise comparisons, allowing a dynamic scheduler, micro-batch aggregation, and weighted Bayesian updates to maintain a stable online leaderboard despite changing prompt distributions and model injection. Experiments with independently sampled prompt streams demonstrate that continually refreshed prompts provide a robust evaluation protocol, reducing the impact of prompt-set-specific tuning. Simulations and ablations further confirm that the proposed ranking framework achieves a strong balance among cold-start convergence, late-entry discovery, and long-run ranking fidelity. |
| title | DynT2I-Eval: A Dynamic Evaluation Framework for Text-to-Image Models |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2605.06170 |