Representation Forcing for Bottleneck-Free Unified Multimodal Models
Paper • 2605.31604 • Published • 60
AI & ML interests
None defined yet.
Recent Activity
View all activity
Papers
The Shape of Addition: Geometric Structures of Arithmetic in Large Language Models
NGM: A Plug-and-Play Training-Free Memory Module for LLMs
wangfuyun
authored a
paper 10 days ago
Automationyw
submitted a
paper to Daily Papers 26 days ago
weiqingXiao
submitted a
paper to Daily Papers about 1 month ago
wangfuyun
authored a
paper about 2 months ago
BradyFU
submitted a
paper to Daily Papers 3 months ago
shijuzhao
submitted a
paper to Daily Papers 4 months ago
wangfuyun
authored a
paper 4 months ago
Post
2247
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
wangfuyun
submitted a
paper to Daily Papers 4 months ago
CraneLu
submitted a
paper to Daily Papers 5 months ago
XiaokunSun
submitted a
paper to Daily Papers 5 months ago
wangfuyun
authored 2
papers 6 months ago
wangfuyun
submitted a
paper to Daily Papers 6 months ago
hehesang
authored a
paper 10 months ago
bigorange111
authored 3
papers 10 months ago
Multi-Prompting Decoder Helps Better Language Understanding
Paper • 2406.06279 • Published
Contrastive Prompting Enhances Sentence Embeddings in LLMs through Inference-Time Steering
Paper • 2505.12831 • Published • 1
Steering When Necessary: Flexible Steering Large Language Models with Backtracking
Paper • 2508.17621 • Published • 2
Swtheking
authored 2
papers about 1 year ago