视觉定位论文趋势 2023–2026

主要结论

• 视觉定位论文从 2023 年的 23 篇增长到 2026 年的 100 篇，三年间扩大到约 4.3 倍。
• CVPR 是发视觉定位论文最多的会议，2023–2026 年共 108 篇，占四个会议总量的 52%。
• 交叉方向里涨得最快的是多模态大模型：同时属于两个主题的论文从 2023 年的 2 篇增至 2026 年的 56 篇。
• 在「开放词汇视觉」的 5 个主题里，它的 2026 年论文数排第 3；同组增长最快的开放词汇检测是 1.5 倍，视觉定位是 4.3 倍。

代表论文

按年份和会议均匀挑选，方便快速看到每一届的样貌

• NeurIPS 2026From Squeezing to Grounding: Visual Guided DPO for Multimodal Hallucination Mitigation
• ICML 2026PGT: Procedurally Generated Tasks for improving visual grounding in MLLMs
• ICLR 2026How Do Medical MLLMs Fail? A Study on Visual Grounding in Medical Images
• CVPR 2026EG-3DVG Expression and Geometry Aware Grounding Decoder for 3D Visual
• NeurIPS 2025From Objects to Anywhere: A Holistic Benchmark for Multi-level Visual Grounding in 3D Scenes
• ICML 2025Fundamental Limits of Visual Autoregressive Transformers: Universal Approximation Abilities
• ICLR 2025DenseGrounding: Improving Dense Language-Vision Semantics for Ego-centric 3D Visual Grounding
• CVPR 2025Discovering Hidden Visual Concepts Beyond Linguistic Input in Infant Learning
• NeurIPS 2024IKEA Manuals at Work: 4D Grounding of Assembly Instructions on Internet Videos
• ICLR 2024CoT3DRef: Chain-of-Thoughts Data-Efficient 3D Visual Grounding
• CVPR 2024EmoVIT: Revolutionizing Emotion Insights with Visual Instruction Tuning
• NeurIPS 2023Grounding Neural Inference with Satisfiability Modulo Theories
• ICML 2023Rethinking Visual Reconstruction: Experience-Based Content Completion Guided by Visual Cues
• ICLR 2023Learning to reason over visual objects
• CVPR 2023Advancing Visual Grounding With Scene Knowledge: Benchmark and Method
• NeurIPS 2026Not All Instances Should Contribute Equally: Sparse Support Grounding via Unbalanced Transport for Heterogeneous MIL
• ICML 2026Grounding Functional Similarity by Invariance-Aware Model Stitching
• ICLR 2026CloDS: Visual-Only Unsupervised Cloth Dynamics Learning in Unknown Conditions
• CVPR 2026Enhancing Part-Level Point Grounding for Any Open-Source MLLMs
• NeurIPS 2025On the rankability of visual embeddings

全部视觉定位论文

共 14 篇，第 1 / 1 页

1. ICML 2026AmbiRefer3D: 3D Visual Grounding with Referential Ambiguity
2. ICML 2026CACR: Reinforcing Temporal Answer Grounding in Instructional Video via Candidate-Aware Causal Reasoning
3. ICML 2026CLIP Tricks You: Training-free Token Pruning for Efficient Pixel Grounding in Large Vision-Language Models
4. ICML 2026Decomposed On-Policy Distillation for Vision-Language Reasoning: Steering Gradients for Visual Grounding
5. ICML 2026DSGCR: Decomposed Spectral Geometry-Aware Cross-Modal Semantic Representation for 3D Visual Grounding
6. ICML 2026Efficient Skill Grounding via Code Refactoring with Small Language Models
7. ICML 2026Grounding Functional Similarity by Invariance-Aware Model Stitching
8. ICML 2026GUI-Spotlight: Adaptive Iterative Focus Refinement for Enhanced GUI Visual Grounding
9. ICML 2026Learning Attribute–Affordance Hierarchies in Hyperbolic Space for Open-Vocabulary 3D Object Affordance Grounding
10. ICML 2026Learning Sparse Visual Representations via Spatial-Semantic Factorization
11. ICML 2026MedSIGHT: Towards Grounded Visual Comprehension in Medical Large Vision-Language Models
12. ICML 2026PGT: Procedurally Generated Tasks for improving visual grounding in MLLMs
13. ICML 2026Rethinking Visual Autoregressive Sampling with Information-Grounding Guidance
14. ICML 2026Unveiling the Visual Counting Bottleneck in Vision-Language Models

常见问题

2023–2026 年四大会议一共有多少篇视觉定位论文？

共 209 篇，按年份分别是 2023 年 23 篇、2024 年 42 篇、2025 年 44 篇、2026 年 100 篇。

NeurIPS 2026 有多少篇视觉定位论文？

NeurIPS 2026 有 33 篇视觉定位论文，上一届是 13 篇。

ICLR 2026 有多少篇视觉定位论文？

ICLR 2026 有 9 篇视觉定位论文，上一届是 9 篇。

ICML 2026 有多少篇视觉定位论文？

ICML 2026 有 14 篇视觉定位论文，上一届是 3 篇。

CVPR 2026 有多少篇视觉定位论文？

CVPR 2026 有 44 篇视觉定位论文，上一届是 19 篇。

哪个会议的视觉定位论文最多？

CVPR，2023–2026 年共 108 篇。