佐藤 智也

Pixel-Leo's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted a paper 2 days ago

Learning Latent Protein Languages for Autoregressive Generation

Paper • 2610.03978 • Published 7 days ago • 6

upvoted a paper 19 days ago

What Does Privileged Information Add to On-Policy Self-Distillation?

Paper • 2609.20612 • Published 22 days ago • 36

upvoted a paper 20 days ago

When EOS Tokens Disagree: Understanding Length Inflation in On-Policy Distillation

Paper • 2609.20511 • Published 22 days ago • 115

upvoted a paper 23 days ago

Continual Learning Mechanisms Compose for Long-Horizon Memorization

Paper • 2609.06986 • Published Sep 7 • 376