22 17 26

Fu-Yun Wang

wangfuyun

https://g-u-n.github.io

g-u-n

AI & ML interests

None yet

Recent Activity

authored a paper about 19 hours ago

PromptRL: Prompt Matters in RL for Flow-Based Image Generation

posted an update about 22 hours ago

PromptRL: Language Models as Co-Learners in Flow-Based Image Generation RL 🚀 We found two critical failure modes in flow-based RL: 1️⃣ Quality-Diversity Dilemma: High-quality models produce similar outputs, bottlenecking RL exploration 2️⃣ Prompt Linguistic Hacking: Models overfit to surface patterns—paraphrase the prompt and performance tanks Solution: **Jointly train LM + FM** — the LM dynamically generates semantically-consistent but diverse prompt variants 📊 Results: • GenEval: 0.97 • OCR accuracy: 0.98 • PickScore: 24.05 • 2×+ fewer rollouts than flow-only RL Paper: arxiv.org/abs/2602.01382 Code: github.com/G-U-N/UniRL #AI #TextToImage #ReinforcementLearning #Diffusion

upvoted an article 1 day ago

PromptRL: Prompt Matters in RL for Flow-Based Image Generation

View all activity

Organizations

authored a paper about 19 hours ago

PromptRL: Prompt Matters in RL for Flow-Based Image Generation

Paper • 2602.01382 • Published 2 days ago • 6

posted an update about 22 hours ago

Post

135

PromptRL: Language Models as Co-Learners in Flow-Based Image Generation RL 🚀

We found two critical failure modes in flow-based RL:
1️⃣ Quality-Diversity Dilemma: High-quality models produce similar outputs, bottlenecking RL exploration
2️⃣ Prompt Linguistic Hacking: Models overfit to surface patterns—paraphrase the prompt and performance tanks

Solution: **Jointly train LM + FM** — the LM dynamically generates semantically-consistent but diverse prompt variants

📊 Results:
• GenEval: 0.97
• OCR accuracy: 0.98
• PickScore: 24.05
• 2×+ fewer rollouts than flow-only RL

Paper: arxiv.org/abs/2602.01382
Code: github.com/G-U-N/UniRL

#AI #TextToImage #ReinforcementLearning #Diffusion