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 11,207 篇，第 16 / 225 页

751. CVPR 2026Multimodal Continual Instruction Tuning with Dynamic Gradient Guidance
752. CVPR 2026Multimodal Distribution Matching for Vision-Language Dataset Distillation
753. CVPR 2026Multimodal Learning on Low-Quality Data with Conformal Predictive Self-Calibration
754. CVPR 2026Multimodal Protein Language Models for Enzyme Kinetic Parameters From Substrate
755. CVPR 2026Multimodal RewardBench 2 Evaluating Omni Reward Models for Interleaved Text
756. CVPR 2026Multimodal Semantic Bias Mitigation for Diverse Text-To-3D Generation
757. CVPR 2026MultiModalPFN Extending Prior-Data Fitted Networks for Multimodal Tabular Learning
758. CVPR 2026MuM Multi-View Masked Image Modeling for 3D Vision
759. CVPR 2026MUSE Harnessing Precise and Diverse Semantics for Few-Shot Whole Slide
760. CVPR 2026Muses Designing Composing Generating Nonexistent Fantasy 3D Creatures without Training
761. CVPR 2026MuViT Multi-Resolution Vision Transformers for Learning Across Scales in Microscopy
762. CVPR 2026MVGGT Multimodal Visual Geometry Grounded Transformer for Multiview 3D Referring
763. CVPR 2026mVLM A Vision Language Model for mNPUs
764. CVPR 2026MVLM Template-Free Tracking via Vision-Language Margin Confidence and Memory-Gated Tracking
765. CVPR 2026NavForesee A Unified Vision-Language World Model for Hierarchical Planning and
766. CVPR 2026NERFIFY A Multi-Agent Framework for Turning NeRF Papers into Code
767. CVPR 2026Neural Differentiation in Deep Networks A Theoretical Framework for Expressivity
768. CVPR 2026Neural-Centric Video Processing Pipeline for Unified Multi-Task Inference
769. CVPR 2026Next-Scale Prediction A Self-Supervised Approach for Real-World Image Denoising
770. CVPR 2026NitroGen An Open Foundation Model for Generalist Gaming Agents
771. CVPR 2026No Hard Negatives Required Concept Centric Learning Leads to Compositionality
772. CVPR 2026No Need For Real Anomaly MLLM Empowered Zero-Shot Video Anomaly
773. CVPR 2026Noise-Aware Few-Shot Learning through Bi-directional Multi-View Prompt Alignment
774. CVPR 2026Nonparametric Deep Fine-grained Clustering with Low-Rank Guided Vision-Language Model
775. CVPR 2026NoOVD Novel Category Discovery and Embedding for Open-Vocabulary Object Detection
776. CVPR 2026NoRD A Data-Efficient Vision-Language-Action Model that Drives without Reasoning
777. CVPR 2026NOVA Sparse Control Dense Synthesis for Pair-Free Video Editing
778. CVPR 2026NOWA Null-space Optical Watermark for Invisible Capture Fingerprinting and Tamper
779. CVPR 2026NuWa Deriving Lightweight Class-Specific Vision Transformers for Edge Devices
780. CVPR 2026OASIS On-Demand Hierarchical Event Memory for Streaming Video Reasoning
781. CVPR 2026Obstruction Reasoning for Robotic Grasping
782. CVPR 2026OctoMed Data Recipes for State-of-the-Art Multimodal Medical Reasoning
783. CVPR 2026OctoNav Towards Generalist Embodied Navigation
784. CVPR 2026Octopus History-Free Gradient Orthogonalization for Continual Learning in Multimodal Large
785. CVPR 2026OddGridBench Exposing the Lack of Fine-Grained Visual Discrepancy Sensitivity in
786. CVPR 2026Olbedo An Albedo and Shading Aerial Dataset for Large-Scale Outdoor
787. CVPR 2026OlmoEarth Stable Latent Image Modeling for Multimodal Earth Observation
788. CVPR 2026Omni-Attack Adversarial Attacks on Open-Ended VQA in Black-Box Multimodal LLMs
789. CVPR 2026Omni-Fake Benchmarking Unified Multimodal Social Media Deepfake Detection
790. CVPR 2026OmniBrainBench A Comprehensive Multimodal Benchmark for Brain Imaging Analysis Across
791. CVPR 2026OmniFM Toward Modality-Robust and Task-Agnostic Federated Learning for Heterogeneous Medical
792. CVPR 2026OmniFood8K Single-Image Nutrition Estimation via Hierarchical Frequency-Aligned Fusion
793. CVPR 2026OmniGen2 Towards Instruction-Aligned Multimodal Generation
794. CVPR 2026OmniGround A Comprehensive Spatio-Temporal Grounding Benchmark for Real-World Complex Scenarios
795. CVPR 2026OmniLottie Generating Vector Animations via Parameterized Lottie Tokens
796. CVPR 2026OmniVGGT Omni-Modality Driven Visual Geometry Grounded Transformer
797. CVPR 2026On Tokens Dilemma Dynamic MoE with Drift-Aware Token Assignment for
798. CVPR 2026One Algorithm to Align Them All
799. CVPR 2026One Layers Trash is Another Layers Treasure Adaptive Layer-wise Visual
800. CVPR 2026One Patch to Caption Them All A Unified Zero-Shot Captioning

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。