Fast and Fluent Diffusion Language Models via Convolutional Decoding and Rejective Fine-tuning

Documenti analoghi

• Train-Attention: Meta-Learning Where to Focus in Continual Knowledge Learning
  di: Seo, Yeongbin, et al.
  Pubblicazione: (2024)
• Quantifying Genuine Awareness in Hallucination Prediction Beyond Question-Side Shortcuts
  di: Seo, Yeongbin, et al.
  Pubblicazione: (2025)
• Prior-based Noisy Text Data Filtering: Fast and Strong Alternative For Perplexity
  di: Seo, Yeongbin, et al.
  Pubblicazione: (2025)
• Efficient Adaptive Rejection Sampling for Accelerating Speculative Decoding in Large Language Models
  di: Sun, Chendong, et al.
  Pubblicazione: (2025)
• WeDLM: Reconciling Diffusion Language Models with Standard Causal Attention for Fast Inference
  di: Liu, Aiwei, et al.
  Pubblicazione: (2025)