Let's Predict Sentence by Sentence
Paper
•
2505.22202
•
Published
•
19
nlp
-
-
From Bytes to Ideas: Language Modeling with Autoregressive U-Nets
Paper • 2506.14761 • Published • 17 -
TokAlign: Efficient Vocabulary Adaptation via Token Alignment
Paper • 2506.03523 • Published -
zip2zip: Inference-Time Adaptive Vocabularies for Language Models via Token Compression
Paper • 2506.01084 • Published • 7 -
Pretraining Language Models to Ponder in Continuous Space
Paper • 2505.20674 • Published • 2