DeepSpeed-VisualChat: Multi-Round Multi-Image Interleave Chat via Multi-Modal Causal Attention
Paper
•
2309.14327
•
Published
•
22
Ankush Collection
Transformer Articles
-
-
MambaVision: A Hybrid Mamba-Transformer Vision Backbone
Paper • 2407.08083 • Published • 30 -
Memory^3: Language Modeling with Explicit Memory
Paper • 2407.01178 • Published • 4 -
Teaching Transformers Causal Reasoning through Axiomatic Training
Paper • 2407.07612 • Published • 2 -
mistralai/Mamba-Codestral-7B-v0.1
Updated • 4.72k • 579 -
PoseLess: Depth-Free Vision-to-Joint Control via Direct Image Mapping with VLM
Paper • 2503.07111 • Published • 2