图像编辑论文趋势 2023–2026
指令编辑
统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 692 篇被归入该主题的论文。
主要结论
- 图像编辑论文从 2023 年的 112 篇增长到 2026 年的 238 篇，三年间扩大到约 2.1 倍。
- CVPR 是发图像编辑论文最多的会议，2023–2026 年共 364 篇，占四个会议总量的 53%。
- 交叉方向里涨得最快的是应用 · computer vision：同时属于两个主题的论文从 2023 年的 1 篇增至 2026 年的 15 篇。
- 在「多模态生成」的 5 个主题里，它的 2026 年论文数排第 3；同组增长最快的叙事生成是 15 倍，图像编辑是 2.1 倍。
各会议论文数
按年份和会议统计
| 年份 | NeurIPS | ICLR | ICML | CVPR | 合计 |
|---|---|---|---|---|---|
| 2023 | 25 | 15 | 8 | 64 | 112 |
| 2024 | 33 | 24 | 13 | 90 | 160 |
| 2025 | 45 | 29 | 11 | 97 | 182 |
| 2026 | 52 | 40 | 33 | 113 | 238 |
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 2026Accelerated Image Editing via Consistency-Aware Source Token Pruning
- ICML 2026Klein Hyperbolic Metric Learning
- ICLR 2026DiffVax: Optimization-Free Image Immunization Against Diffusion-Based Editing
- CVPR 2026AnchorFlow Training-Free 3D Editing via Latent Anchor-Aligned Flows
- NeurIPS 2025A$^3$E: Towards Compositional Model Editing
- ICML 2025DEALing with Image Reconstruction: Deep Attentive Least Squares
- ICLR 2025HQ-Edit: A High-Quality Dataset for Instruction-based Image Editing
- CVPR 2025AnyEdit: Mastering Unified High-Quality Image Editing for Any Idea
- NeurIPS 2024Reproducibility study of “LICO: Explainable Models with Language-Image Consistency"
- ICML 2024Counterfactual Image Editing
- ICLR 2024Guiding Instruction-based Image Editing via Multimodal Large Language Models
- CVPR 2024DiffEditor: Boosting Accuracy and Flexibility on Diffusion-based Image Editing
- NeurIPS 2023ImageBrush: Learning Visual In-Context Instructions for Exemplar-Based Image Manipulation
- ICML 2023Random Shuffle Transformer for Image Restoration
- ICLR 2023Editing models with task arithmetic
- CVPR 20233D GAN Inversion With Facial Symmetry Prior
- NeurIPS 2026AnyBokeh: Physics-Guided Any-to-Any Bokeh Editing with Optical Fingerprint Transfer
- ICML 2026DLEBench: Evaluating Small-scale Object Editing Ability for Instruction-based Image Editing Model
- ICLR 2026Energy-Regularized Sequential Model Editing on Hyperspheres
- CVPR 2026ChordEdit One-Step Low-Energy Transport for Image Editing
全部图像编辑论文
共 33 篇，第 1 / 1 页
- ICML 2026AdaEraser: Training-Free Object Removal via Adaptive Attention Suppression
- ICML 2026AdLift: Lifting Adversarial Perturbations to Safeguard 3D Gaussian Splatting Assets Against Instruction-Driven Editing
- ICML 2026Alterbute: Editing Intrinsic Attributes of Objects in Images
- ICML 2026B-Spar: Bayesian Sparse-Reward Modeling for RL-based Image Editing
- ICML 2026CacheEdit: Efficient Multi-round Image Editing via Adaptive Token-wise Reuse.
- ICML 2026ChartE$^{3}$: A Comprehensive Benchmark for End-to-End Chart Editing
- ICML 2026CoCoEdit: Content-Consistent Image Editing via Region Regularized Reinforcement Learning
- ICML 2026DirectEdit: Step-Level Accurate Inversion for Flow-Based Image Editing
- ICML 2026DLEBench: Evaluating Small-scale Object Editing Ability for Instruction-based Image Editing Model
- ICML 2026Entropy-Aware Dynamic KV Cache Sparsification for Autoregressive Image Generation and Editing
- ICML 2026From Backward Spreading to Forward Replay: Revisiting Target Construction in LLM Parameter Editing
- ICML 2026From Statics to Dynamics: Physics-Aware Image Editing with Latent Transition Priors
- ICML 2026Geometrically Constrained Stenosis Editing in Coronary Angiography via Entropic Optimal Transport
- ICML 2026Image Restoration via Diffusion Models with Dynamic Resolution
- ICML 2026Klein Hyperbolic Metric Learning
- ICML 2026MICE-Bench: A Challenging and Comprehensive Benchmark for Multi-Reference Image Creation and Editing
- ICML 2026MiVE: Multiscale Vision-language features for reference-guided video Editing
- ICML 2026Order within Chaos: Capturing Intrinsic Energy Anomalies for AI-Manipulated Image Forgery Localization
- ICML 2026Reasoning to Edit: Hypothetical Instruction-Based Image Editing with Visual Reasoning
- ICML 2026RevealLayer: Disentangling Hidden and Visible Layers via Occlusion-Aware Image Decomposition
- ICML 2026Semantic Editing with Coupled Stochastic Differential Equations
- ICML 2026Semantic Granularity Navigation in Image Editing
- ICML 2026Shifting the Breaking Point of Flow Matching for Multi-Instance Editing
- ICML 2026SlerpFlow: Spherical Trajectory Correction for Rectified Flow Inversion
- ICML 2026Sparse Relaxed-Lasso Steering: Automatic Sparse Autoencoder Feature Selection for Precise Image Editing
- ICML 2026StableI2I: Spotting Unintended Changes in Image-to-Image Transition
- ICML 2026StyleDistillation: A New Insight of Image Style Enables Personalized Aesthetic Manipulation
- ICML 2026T-Edit: Triple-Branch Diffusion Anchoring for Consistent Editing
- ICML 2026Test-Time Anchoring for Discrete Diffusion Posterior Sampling
- ICML 2026Towards Diffeomorphism-Equivariant Neural Networks via Canonicalization
- ICML 2026TransLight: Image-Guided Customized Lighting Control with Generative Decoupling
- ICML 2026Unpaired Visual Editing with Self-Consistent Flow Matching
- ICML 2026When the Prompt Becomes Visual: Vision-Centric Jailbreak Attacks for Large Image Editing Models
在 Claude、Cursor 里直接问这个趋势
接上 Paperlist MCP 后，写综述、挑方向或准备投稿时，不用离开对话框就能拿到这页的数字：每年各会议的论文数、增长倍数、涨得最快的子方向和代表论文，AI 会按你的问题整理。
可以直接复制这句话问 AI
用 paperlist 查图像编辑在 CVPR 2023–2026 的论文趋势，哪一年涨得最快？
AI 会调用的工具：get_topic_stats {"topic": "image-editing"}
常见问题
- 2023–2026 年四大会议一共有多少篇图像编辑论文？
- 共 692 篇，按年份分别是 2023 年 112 篇、2024 年 160 篇、2025 年 182 篇、2026 年 238 篇。
- NeurIPS 2026 有多少篇图像编辑论文？
- NeurIPS 2026 有 52 篇图像编辑论文，上一届是 45 篇。
- ICLR 2026 有多少篇图像编辑论文？
- ICLR 2026 有 40 篇图像编辑论文，上一届是 29 篇。
- ICML 2026 有多少篇图像编辑论文？
- ICML 2026 有 33 篇图像编辑论文，上一届是 11 篇。
- CVPR 2026 有多少篇图像编辑论文？
- CVPR 2026 有 113 篇图像编辑论文，上一届是 97 篇。
- 哪个会议的图像编辑论文最多？
- CVPR，2023–2026 年共 364 篇。