Research map / Vision-language and multimodal / Multimodal LLMs
Toward Visual Grounding: A Survey
Linhui Xiao, Xiaoshan Yang, Xiangyuan Lan, Yaowei Wang, Changsheng Xu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace unimodal detector/BERT backbone with VLP model (BEiT3) | single-dataset fine-tuning · RefCOCO val | accuracy: 86.28 -> 88.75 | Improved |
| add dataset-mixed intermediate grounding pre-training | OneRef · RefCOCO val | accuracy: 88.75 -> 91.89 | Improved |
| add dataset-mixed intermediate grounding pre-training | HiVG · RefCOCO val | accuracy: 87.32 -> 90.56 | Improved |
| replace one-stage convolutional detector with Transformer (TransVG) | visual grounding architecture · RefCOCO val | accuracy: 79.67 -> 81.02 | Improved |
| use fine-tuned grounding MLLM instead of unimodal detector Transformer | grounding model · RefCOCO val | accuracy: 81.02 -> 87.01 | Mixed |
| retrain on train, val and test sets to estimate upper limit | OneRef-B · RefCOCO val | accuracy: 91.89 -> 98.69 | Improved |
| retrain on train, val and test sets to estimate upper limit | OneRef-B · RefCOCO testB | accuracy: 88.58 -> 98.57 | Improved |
| use VLP-based transfer instead of proposal-based weak supervision | PPT vs PSRN · RefCOCO+ val | accuracy: 40.68 -> 68.16 | Improved |
| use one-stage CLIP-based weak supervision instead of proposal-based | QueryMatch vs PSRN · RefCOCO+ val | accuracy: 40.68 -> 44.76 | Improved |
Closest papers by meaning
- GroundingME: Exposing the Visual Grounding Gap in MLLMs through Multi-Dimensional Evaluation (CVPR 2026 · similarity 0.92)
- SimVG: A Simple Framework for Visual Grounding with Decoupled Multi-modal Fusion (NeurIPS 2024 · similarity 0.92)
- Your Large Vision-Language Model Only Needs A Few Attention Heads For Visual Grounding (CVPR 2025 · similarity 0.91)
- GeM-VG: Towards Generalized Multi-image Visual Grounding with Multimodal Large Language Models (AAAI 2026 · similarity 0.91)
- Referencing Where to Focus: Improving Visual Grounding with Referential Query (NeurIPS 2024 · similarity 0.91)
- Connecting the Dots: Training-Free Visual Grounding via Agentic Reasoning (AAAI 2026 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
