APT coordinates three agents.
Adaptive feedback targets each image model’s changing weaknesses
APT uses three agents to evaluate a text-to-image model, diagnose failures and generate model-specific reinforcement-learning prompts.
Big Tech
Sirui Zhang · Fangyun Wei · Dong Chen · Jinjing Zhao · Yitong Wang · Yunuo Chen · +2 more
University of Science and Technology of China · Microsoft Research Asia · University of Sydney · Fudan University · Shanghai Jiao Tong University
Research Digest··3 min read
Zhang et al.
Why this paper
From Microsoft Research Asia and 4 others
In one line
Adaptive, feedback-driven post-training improves diverse text-to-image models more than handcrafted reinforcement-learning data by targeting each model’s evolving weaknesses.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ·No benchmark numbers found
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§