Fast and Fluent Diffusion Language Models via Convolutional Decoding and Rejective Fine-tuning
Paper • 2509.15188 • Published • 1
Yeongbin Seo
YBSeo1
AI & ML interests
LLM, diffusion LM, hierarchical LM
Recent Activity
authored a paper 3 days ago
Fast and Fluent Diffusion Language Models via Convolutional Decoding and Rejective Fine-tuning authored a paper 3 days ago
Train-Attention: Meta-Learning Where to Focus in Continual Knowledge Learning authored a paper 3 days ago
Prior-based Noisy Text Data Filtering: Fast and Strong Alternative For PerplexityOrganizations
None yet
authored 4
papers 3 days ago
Train-Attention: Meta-Learning Where to Focus in Continual Knowledge Learning
Paper • 2407.16920 • Published • 1
Prior-based Noisy Text Data Filtering: Fast and Strong Alternative For Perplexity
Paper • 2509.18577 • Published • 1
Quantifying Genuine Awareness in Hallucination Prediction Beyond Question-Side Shortcuts
Paper • 2509.15339 • Published • 1