Differentiable Virtual-Slot Partitioning for Adaptive Tokenization in Vision Transformers
cs.CV
Submitted: 2026-11-08
Updated: 2026-11-08
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Vision Transformers with Mixed-Resolution Tokenization
- DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification
- DART: Differentiable Dynamic Adaptive Region Tokenizer for Vision Foundation Models
- AdaViT: Adaptive Tokens for Efficient Vision Transformer
- Not All Images are Worth 16x16 Words: Dynamic Transformers for Efficient Image Recognition
- Conditional Positional Encodings for Vision Transformers
- SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers
- EVA-02: A Visual Representation for Neon Genesis
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs