Jag: Direct Box Prediction for Efficient Visual Grounding
cs.CV
Submitted: 2026-10-09
Updated: 2026-10-12
Comments: 16 pages, 6 figures, 13 tables
Project page: https://xyzzzh.github.io/Jag
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond
- Qwen3-VL Technical Report
- Qwen2.5-VL Technical Report
- Revisiting Referring Expression Comprehension Evaluation in the Era of Large Multimodal Models
- Shikra: Unleashing Multimodal LLM's Referential Dialogue Magic
- Detect Anything via Next Point Prediction
- Visual Instruction Tuning
- Visual-RFT: Visual Reinforcement Fine-Tuning
- VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model
- LocateAnything: Fast and High-Quality Vision-Language Grounding with Parallel Box Decoding
- DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding
- SWIFT:A Scalable lightWeight Infrastructure for Fine-Tuning
- InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models
- Hi-Token: Hierarchical Coordinate Tokenization for Generative Visual Grounding
- IoU-PD: IoU-Aware Privileged Distillation for Visual Grounding with Multimodal Large Language Models
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models