wangfuyun/PrompRL
Updated • 2
Fu-Yun Wang
wangfuyun
AI & ML interests
None yet
Recent Activity
updated a model about 5 hours ago
wangfuyun/PrompRL authored a paper 24 days ago
Context Unrolling in Omni Models upvoted a paper 27 days ago
Context Unrolling in Omni ModelsOrganizations
updated a
model about 5 hours ago
authored a
paper 24 days ago
upvoted a paper 27 days ago
liked a
model about 1 month ago
reacted to their post with 🚀 4 months ago
Post
2123
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
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
authored a
paper 4 months ago
posted an update 4 months ago
Post
2123
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
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 4 months ago
Article
PromptRL: Prompt Matters in RL for Flow-Based Image Generation
wangfuyun
• • 3 published an article 4 months ago
Article
PromptRL: Prompt Matters in RL for Flow-Based Image Generation
wangfuyun
• • 3 liked a
model 4 months ago
upvoted a paper 4 months ago
submitted a
paper to Daily Papers 4 months ago
liked a
model 4 months ago
published 2
models 4 months ago
updated a
model 4 months ago
liked a
model 5 months ago
updated a
model 5 months ago
liked a
model 5 months ago
published a
model 5 months ago