Visual Token Routing: Learning to keep tokens that matter
cs.CV, cs.LG
Submitted: 2026-11-04
Updated: 2026-11-04
Project page: https://pranjal2041.github.io/vtr
Terminology
Sources
- ReVision: Scaling Computer-Use Agents via Temporal Visual Redundancy Reduction
- Gym-Anything-RT: Scaling Real-Time Environments for Vision-Language Agents
- cua-speedrun: Standardized Benchmarking of the Speed of Computer-Use Agents
- Gym-Anything: Turn any Software into an Agent Environment
- Efficient Video Sampling: Pruning Temporally Redundant Tokens for Faster VLM Inference
- Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation
- Adaptive Computation Time for Recurrent Neural Networks
- UI-TARS: Pioneering Automated GUI Interaction with Native Agents
- Mixture-of-Depths: Dynamically allocating compute in transformer-based language models
- Taming VAEs
- CoPE-VideoLM: Leveraging Codec Primitives For Efficient Video Language Modeling
- OneVision-Encoder: Codec-Aligned Sparsity as a Foundational Principle for Multimodal Intelligence
- Dynamic Token Compression for Efficient Video Understanding through Reinforcement Learning
- STP-GUI: Spatially-Aware Token Pruning for Efficient Web-GUI Agents
- Spatio-Temporal Token Pruning for Efficient High-Resolution GUI Agents
- VideoGameBench: Can Vision-Language Models complete popular video games?
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