多模态大模型论文趋势 2023–2026

主要结论

• 多模态大模型论文从 2023 年的 1,226 篇增长到 2026 年的 5,192 篇，三年间扩大到约 4.2 倍。
• NeurIPS 是发多模态大模型论文最多的会议，2023–2026 年共 3,319 篇，占四个会议总量的 30%。
• 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 35 篇增至 2026 年的 596 篇。
• 在「前沿系统」的 4 个主题里，它的 2026 年论文数排第 1；同组增长最快的AI 智能体是 12 倍，多模态大模型是 4.2 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 2026$\text{A}^3$-VLA: Automatic Perception-Guided Attention Alignment for Robot Manipulation
• ICML 2026[CLS] is Not Enough: Multi-Label Recognition via Patch-Level Inference and Adaptive Aggregation
• ICLR 2026A Balanced Neuro-Symbolic Approach for Commonsense Abductive Logic
• CVPR 20263D Space as a Scratchpad for Editable Text-to-Image Generation
• NeurIPS 2025$\Delta \mathrm{Energy}$: Optimizing Energy Change During Vision-Language Alignment Improves both OOD Detection and OOD Generalization
• ICML 2025Adapting While Learning: Grounding LLMs for Scientific Problems with Tool Usage Adaptation
• ICLR 2025$\gamma-$MoD: Exploring Mixture-of-Depth Adaptation for Multimodal Large Language Models
• CVPR 20252DMamba: Efficient State Space Model for Image Representation with Applications on Giga-Pixel Whole Slide Image Classification
• NeurIPS 2024A Concept-Based Explainability Framework for Large Multimodal Models
• ICML 2024A Multimodal Automated Interpretability Agent
• ICLR 2024Alt-Text with Context: Improving Accessibility for Images on Twitter
• CVPR 2024Data-Efficient Multimodal Fusion on a Single GPU
• NeurIPS 20234M: Massively Multimodal Masked Modeling
• ICML 2023$\pi$-Tuning: Transferring Multimodal Foundation Models with Optimal Multi-task Interpolation
• ICLR 2023A Primal-Dual Framework for Transformers and Neural Networks
• CVPR 20233D Concept Learning and Reasoning From Multi-View Images
• NeurIPS 2026$\texttt{bispectrum}$: Selective $G$-Bispectra Made Practical
• ICML 2026$G^2$-Reader: Dual Evolving Graphs for Multimodal Document QA
• ICLR 2026A Comprehensive Information-Decomposition Analysis of Large Vision-Language Models
• CVPR 20263D-IDE 3D Implicit Depth Emergent

全部多模态大模型论文

共 3,319 篇，第 1 / 67 页

1. NeurIPS 2026$\alpha$Depth: Learning Single-Pass Soft Boundary Decomposition for Stereo Conversion
2. NeurIPS 2026$\text{A}^3$-VLA: Automatic Perception-Guided Attention Alignment for Robot Manipulation
3. NeurIPS 2026$\texttt{bispectrum}$: Selective $G$-Bispectra Made Practical
4. NeurIPS 2026$h$-control: Training-Free Camera Control via Block-Conditional Gibbs Refinement
5. NeurIPS 20262D Spatial Reasoning with Adaptive Neural Cellular Automata
6. NeurIPS 20263A-VLA: Abstraction-Aligned Action Learning for Vision-Language Agents in 3D Game Worlds
7. NeurIPS 20263D-PLOT-LLM: Part-Level Object Tokens for 3D Large Language Models
8. NeurIPS 20263DCodeBench: Benchmarking Agentic Procedural 3D Modeling Via Code
9. NeurIPS 20263DSPA: A 3D Semantic Point Autoencoder for Evaluating Video Realism
10. NeurIPS 2026A Benchmark for Omni-Modal Reasoning in Long Videos
11. NeurIPS 2026A Bitter Lesson for Data Filtering
12. NeurIPS 2026A Closer Look at Dynamic Scene Graph Generation in the Era of Multimodal Large Language Models
13. NeurIPS 2026A Composite Activation Function for Learning Stable Binary Representations
14. NeurIPS 2026A Deep Learning Framework for Scalar-on-Function Models
15. NeurIPS 2026A Diagnostic Benchmark for Layered Layout and Template-Variant Reasoning
16. NeurIPS 2026A Dynamic Decomposition Strategy for the MOEA/D With Proven Performance Guarantees
17. NeurIPS 2026A Favorable Regime Between ODE and SDE for Few-Step Diffusion Sampling
18. NeurIPS 2026A Foundational Model System for Datacenter Machine Repairs
19. NeurIPS 2026A General Concept-based Decomposition for Vision–Language Embeddings
20. NeurIPS 2026A Graph Foundation Model for Unified Clustering
21. NeurIPS 2026A Latent World-Action Model with Jointly Aligned Reasoning
22. NeurIPS 2026A Mechanistic Analysis of Looped Reasoning Language Models
23. NeurIPS 2026A Mechanistic Interpretability Study of an Astronomical Foundation Model
24. NeurIPS 2026A Method-Class Divide in Sub-4-Bit Quantization:\\Iterative vs.\ Single-Pass Sensitivity to Data Composition
25. NeurIPS 2026A Multimodal Benchmark for Evaluating Cause-of-Death Inference Using Child Health and Mortality Data
26. NeurIPS 2026A Principled Optimal Transport Framework for Frame Selection in Long Video Understanding
27. NeurIPS 2026A Provably Convergent and Practical Algorithm for Gromov–Wasserstein Optimal Transport
28. NeurIPS 2026A Scalable Measure of Loss Landscape Curvature for Analyzing the Training Dynamics of LLMs
29. NeurIPS 2026A second order regret bound for NormalHedge
30. NeurIPS 2026A Steerable Deep Network for Model-Free Diffusion MRI Registration
31. NeurIPS 2026A systematic evaluation of vision-language models for observational astronomical reasoning tasks
32. NeurIPS 2026A Theoretical Framework for Self-Play Theorem Proving Algorithms
33. NeurIPS 2026A Theory of Training Profit-Optimal LLMs
34. NeurIPS 2026A Unified Image and Video Encoder for Multimodal LLMs
35. NeurIPS 2026A Unified Spectral Theory of Multimodal Losses
36. NeurIPS 2026A Unifying View of Anchoring via Operator-Side Tikhonov Regularization
37. NeurIPS 2026A$^3$Bench: A Benchmark for Compositional Reasoning over Aggressive Interactions in Videos
38. NeurIPS 2026A2Eval: Agentic and Automated Evaluation for Embodied Brain
39. NeurIPS 2026A2I: Adjacency-to-Image Structural Encodings for Graph Learning
40. NeurIPS 2026Achieve Latency-Efficient Temporal-Coding Spiking LLMs via Discretization-Aware Conversion
41. NeurIPS 2026ActionUNet: Improving Robustness of VLA Models with Efficient Multi-scale Fine-tuning
42. NeurIPS 2026ActQuant: Sub-4-bit Action-Guided Quantization for Vision-Language-Action Models
43. NeurIPS 2026AdaCal: Adaptive Calibration for Robust Sparse Attention in Long-Context LLMs
44. NeurIPS 2026AdaCodec: A Predictive Visual Code for Video MLLMs
45. NeurIPS 2026Adapting Vision Transformers to Organoid Imaging
46. NeurIPS 2026Adaptive Gated Simplicial Propagation for Node Classification in Multimodal Graphs
47. NeurIPS 2026Adaptive Internal Readout for Native Multimodal Models
48. NeurIPS 2026Adaptive Mass-Segmented KV Compression for Long-Context Reasoning
49. NeurIPS 2026Adaptive-Margin Masking and Restoration for Balanced Multimodal Learning
50. NeurIPS 2026AdDirector: Anchored Guidance for Generating Camera-Controllable Advertisement Videos

常见问题

2023–2026 年四大会议一共有多少篇多模态大模型论文？

共 11,209 篇，按年份分别是 2023 年 1,226 篇、2024 年 1,716 篇、2025 年 3,075 篇、2026 年 5,192 篇。

NeurIPS 2026 有多少篇多模态大模型论文？

NeurIPS 2026 有 1,366 篇多模态大模型论文，上一届是 1,074 篇。

ICLR 2026 有多少篇多模态大模型论文？

ICLR 2026 有 1,401 篇多模态大模型论文，上一届是 816 篇。

ICML 2026 有多少篇多模态大模型论文？

ICML 2026 有 1,051 篇多模态大模型论文，上一届是 390 篇。

CVPR 2026 有多少篇多模态大模型论文？

CVPR 2026 有 1,374 篇多模态大模型论文，上一届是 795 篇。

哪个会议的多模态大模型论文最多？

NeurIPS，2023–2026 年共 3,319 篇。