A Frame is Worth One Token: Efficient Generative World Modeling with Delta Tokens
Paper • 2604.04913 • Published • 2
AI & ML interests
Scalable Artificial Intelligence
Recent Activity
View all activity
Papers
A Frame is Worth One Token: Efficient Generative World Modeling with Delta Tokens
Scalable Prompt Routing via Fine-Grained Latent Task Discovery
nielsr
submitted a
paper to Daily Papers about 17 hours ago
LahariChowtoori
updated a
Space 1 day ago
zancato
published 8
models 7 days ago
amazon/Mamba2-primed-HQwen3-8B-Instruct
Text Generation • 8B • Updated • 867 • 1
amazon/GKA-primed-HQwen3-8B-Instruct
Text Generation • 9B • Updated • 912 • 1
amazon/GDN-primed-HQwen3-8B-Instruct
Text Generation • 8B • Updated • 847 • 1
amazon/BMOJOF-primed-HQwen3-8B-Instruct
Text Generation • 9B • Updated • 813 • 1
amazon/GKA-primed-HQwen3-32B-Instruct
Text Generation • 34B • Updated • 832 • 1
amazon/GKA-primed-HQwen3-32B-Reasoner
Text Generation • 34B • Updated • 729 • 1
amazon/GDN-primed-HQwen3-8B-Reasoner
Text Generation • 8B • Updated • 730 • 1
amazon/GKA-primed-HQwen3-8B-Reasoner
Text Generation • 9B • Updated • 788 • 1
zancato
updated 9
models 7 days ago
amazon/GDN-primed-HQwen3-32B-Instruct
Text Generation • 34B • Updated • 841 • 2
amazon/Mamba2-primed-HQwen3-8B-Instruct
Text Generation • 8B • Updated • 867 • 1
amazon/GKA-primed-HQwen3-8B-Instruct
Text Generation • 9B • Updated • 912 • 1
amazon/GDN-primed-HQwen3-8B-Instruct
Text Generation • 8B • Updated • 847 • 1
amazon/BMOJOF-primed-HQwen3-8B-Instruct
Text Generation • 9B • Updated • 813 • 1
amazon/GKA-primed-HQwen3-32B-Instruct
Text Generation • 34B • Updated • 832 • 1
amazon/GKA-primed-HQwen3-32B-Reasoner
Text Generation • 34B • Updated • 729 • 1
amazon/GDN-primed-HQwen3-8B-Reasoner
Text Generation • 8B • Updated • 730 • 1
amazon/GKA-primed-HQwen3-8B-Reasoner
Text Generation • 9B • Updated • 788 • 1
zancato
updated a
collection 8 days ago