指代表达论文趋势 2023–2026

主要结论

• 指代表达论文从 2023 年的 18 篇增长到 2026 年的 27 篇，三年间扩大到约 1.5 倍。
• CVPR 是发指代表达论文最多的会议，2023–2026 年共 59 篇，占四个会议总量的 64%。
• 交叉方向里涨得最快的是三维视觉：同时属于两个主题的论文从 2023 年的 0 篇增至 2026 年的 7 篇。
• 在「开放词汇视觉」的 5 个主题里，它的 2026 年论文数排第 5；同组增长最快的视觉定位是 4.3 倍，指代表达是 1.5 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 2026EviSAM3: Evidence-Driven SAM3 for Referring Remote Sensing Image Segmentation
• ICML 2026Position Is All You Need: A Free Lunch Token Compression Strategy for MLLM-based Referring Expression Segmentation
• ICLR 2026Ref-Adv: Exploring MLLM Visual Reasoning in Referring Expression Tasks
• CVPR 2026CLEX Complementary Label Exchange Learning for Noisy Facial Expression Recognition
• NeurIPS 2025Learning Relative Gene Expression Trends from Pathology Images in Spatial Transcriptomics
• ICML 2025ReferSplat: Referring Segmentation in 3D Gaussian Splatting
• ICLR 2025LLM-wrapper: Black-Box Semantic-Aware Adaptation of Vision-Language Models for Referring Expression Comprehension
• CVPR 2025Decoupled Motion Expression Video Segmentation
• NeurIPS 2024A Large-Scale Human-Centric Benchmark for Referring Expression Comprehension in the LMM Era
• ICML 2024SAM as the Guide: Mastering Pseudo-Label Refinement in Semi-Supervised Referring Expression Segmentation
• ICLR 2024BarLeRIa: An Efficient Tuning Framework for Referring Image Segmentation
• CVPR 2024Decoupling Static and Hierarchical Motion Perception for Referring Video Segmentation
• NeurIPS 2023Segment Everything Everywhere All at Once
• CVPR 2023Contrastive Grouping With Transformer for Referring Image Segmentation
• NeurIPS 2026LeCellModel: Interpretable Density Estimation over the Gene Expression Manifold
• ICML 2026ViSurf: Visual Supervised-and-Reinforcement Fine-Tuning for Large Vision-and-Language Models
• ICLR 2026Patch-as-Decodable-Token: Towards Unified Multi-Modal Vision Tasks in MLLMs
• CVPR 2026InterRVOS Interaction-Aware Referring Video Object Segmentation
• NeurIPS 2025RESAnything: Attribute Prompting for Arbitrary Referring Segmentation
• ICLR 2025Text4Seg: Reimagining Image Segmentation as Text Generation

全部指代表达论文

共 92 篇，第 2 / 2 页

51. CVPR 2024LASO: Language-guided Affordance Segmentation on 3D Object
52. CVPR 2024LoSh: Long-Short Text Joint Prediction Network for Referring Video Object Segmentation
53. CVPR 2024LQMFormer: Language-aware Query Mask Transformer for Referring Image Segmentation
54. CVPR 2024Mask Grounding for Referring Image Segmentation
55. CVPR 2024MirageRoom: 3D Scene Segmentation with 2D Pre-trained Models by Mirage Projection
56. CVPR 2024Mudslide: A Universal Nuclear Instance Segmentation Method
57. CVPR 2024NeRF-HuGS: Improved Neural Radiance Fields in Non-static Scenes Using Heuristics-Guided Segmentation
58. CVPR 2024Prompt-Driven Referring Image Segmentation with Instance Contrasting
59. CVPR 2024Referring Expression Counting
60. CVPR 2024Revisiting Counterfactual Problems in Referring Expression Comprehension
61. CVPR 2024Rotated Multi-Scale Interaction Network for Referring Remote Sensing Image Segmentation
62. CVPR 2024ScanFormer: Referring Expression Comprehension by Iteratively Scanning
63. CVPR 2024Universal Segmentation at Arbitrary Granularity with Language Instruction
64. CVPR 2024Unveiling Parts Beyond Objects: Towards Finer-Granularity Referring Expression Segmentation
65. CVPR 2024Zero-shot Referring Expression Comprehension via Structural Similarity Between Images and Captions
66. ICLR 2024BarLeRIa: An Efficient Tuning Framework for Referring Image Segmentation
67. ICLR 2024EQA-MX: Embodied Question Answering using Multimodal Expression
68. ICML 2024SAM as the Guide: Mastering Pseudo-Label Refinement in Semi-Supervised Referring Expression Segmentation
69. NeurIPS 2024A Large-Scale Human-Centric Benchmark for Referring Expression Comprehension in the LMM Era
70. NeurIPS 2024Boosting Weakly Supervised Referring Image Segmentation via Progressive Comprehension
71. NeurIPS 2024OneRef: Unified One-tower Expression Grounding and Segmentation with Mask Referring Modeling
72. NeurIPS 2024RG-SAN: Rule-Guided Spatial Awareness Network for End-to-End 3D Referring Expression Segmentation
73. NeurIPS 2024SimVG: A Simple Framework for Visual Grounding with Decoupled Multi-modal Fusion
74. NeurIPS 2024Unleashing Region Understanding in Intermediate Layers for MLLM-based Referring Expression Generation
75. CVPR 2023A Soma Segmentation Benchmark in Full Adult Fly Brain
76. CVPR 2023Camouflaged Instance Segmentation via Explicit De-Camouflaging
77. CVPR 2023Contrastive Grouping With Transformer for Referring Image Segmentation
78. CVPR 2023Few-Shot Referring Relationships in Videos
79. CVPR 2023GRES: Generalized Referring Expression Segmentation
80. CVPR 2023Label-Free Liver Tumor Segmentation
81. CVPR 2023Learning To Segment Every Referring Object Point by Point
82. CVPR 2023Meta Compositional Referring Expression Segmentation
83. CVPR 2023Micron-BERT: BERT-Based Facial Micro-Expression Recognition
84. CVPR 2023PolyFormer: Referring Image Segmentation As Sequential Polygon Generation
85. CVPR 2023RefCLIP: A Universal Teacher for Weakly Supervised Referring Expression Comprehension
86. CVPR 2023Referring Image Matting
87. CVPR 2023RefTeacher: A Strong Baseline for Semi-Supervised Referring Expression Comprehension
88. CVPR 2023Rethinking the Learning Paradigm for Dynamic Facial Expression Recognition
89. CVPR 2023Zero-Shot Referring Image Segmentation With Global-Local Context Features
90. NeurIPS 2023Described Object Detection: Liberating Object Detection with Flexible Expressions
91. NeurIPS 2023Fine-Grained Visual Prompting
92. NeurIPS 2023Segment Everything Everywhere All at Once

常见问题

2023–2026 年四大会议一共有多少篇指代表达论文？

共 92 篇，按年份分别是 2023 年 18 篇、2024 年 29 篇、2025 年 18 篇、2026 年 27 篇。

NeurIPS 2026 有多少篇指代表达论文？

NeurIPS 2026 有 6 篇指代表达论文，上一届是 5 篇。

ICLR 2026 有多少篇指代表达论文？

ICLR 2026 有 5 篇指代表达论文，上一届是 2 篇。

ICML 2026 有多少篇指代表达论文？

ICML 2026 有 2 篇指代表达论文，上一届是 1 篇。

CVPR 2026 有多少篇指代表达论文？

CVPR 2026 有 14 篇指代表达论文，上一届是 10 篇。

哪个会议的指代表达论文最多？

CVPR，2023–2026 年共 59 篇。