Continuous Autoregressive Language Models
Paper • 2510.27688 • Published • 75
Tokenization methods and language modelling
-
-
Efficient Speech Language Modeling via Energy Distance in Continuous Latent Space
Paper • 2505.13181 • Published • 9 -
Long-Context Autoregressive Video Modeling with Next-Frame Prediction
Paper • 2503.19325 • Published • 73 -
Bridging Continuous and Discrete Tokens for Autoregressive Visual Generation
Paper • 2503.16430 • Published • 34 -
From Bytes to Ideas: Language Modeling with Autoregressive U-Nets
Paper • 2506.14761 • Published • 17 -
Towards Scalable Pre-training of Visual Tokenizers for Generation
Paper • 2512.13687 • Published • 108 -
Bridging Semantic and Kinematic Conditions with Diffusion-based Discrete Motion Tokenizer
Paper • 2603.19227 • Published • 42 -
TIDE: Every Layer Knows the Token Beneath the Context
Paper • 2605.06216 • Published • 9 -
NITP: Next Implicit Token Prediction for LLM Pre-training
Paper • 2605.24956 • Published • 32 -
OmniPack: Unified Token Compression for Efficient Omni-modal Large Language Models
Paper • 2608.03812 • Published • 17 -
Full-bandwidth transformer
Paper • 2608.08888 • Published • 17 -
G-STAR: Group-Scaled Token Capacity for Deeply Compressed Autoregressive Generation
Paper • 2611.13004 • Published • 37