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 1,366 篇，第 13 / 28 页

601. NeurIPS 2026I-Perceive: A Foundation Model for Vision-Language Active Perception
602. NeurIPS 2026IDEAFix: Evaluation Framework for Creative Defixation Prompting in LLMs
603. NeurIPS 2026IGGT4D: Streaming 4D Instance-Grounded Geometry Transformer
604. NeurIPS 2026Illusory Pattern Perception Drives Spurious Inference in Large Language Models
605. NeurIPS 2026ImmuVis: Hyperconvolutional Foundation Models for Imaging Mass Cytometry
606. NeurIPS 2026Improving Diffusion Posterior Samplers with Lagged Temporal Corrections for Image Restoration
607. NeurIPS 2026Improving Guidance-Free Visual Generation via Self-Contrastive Alignment for Likelihood Estimation
608. NeurIPS 2026Improving Self-Supervised Vision Transformers with Cross Distillation
609. NeurIPS 2026In STeP: Speculative Tensor Parallelism for Concurrent Heterogeneous Inference of LLMs
610. NeurIPS 2026In-Context Learning Can Help Vision Language Models Overcome Training Prior
611. NeurIPS 2026Incentivizing Medical Vision Capabilities from Large-Scale Multimodal Pre-training
612. NeurIPS 2026InduceKV: Fixed-Footprint Continual Adaptation of Multimodal LLMs via Inducing KV Memories
613. NeurIPS 2026InfiniteVL: A Systematic Approach to Highly-Efficient, Ultra-Long Multimodal Understanding
614. NeurIPS 2026InfoNav: A Unified Value Framework Integrating Semantic Relevance and Information Gain for Zero-Shot Object Goal Navigation
615. NeurIPS 2026INFUSER: Influence-Guided Self-Evolution Improves Reasoning
616. NeurIPS 2026Inline Critic Steers Image Editing
617. NeurIPS 2026Innocuous-Seeming Data, Latent Ideology: Ideological Generalisation in Finetuned LLMs
618. NeurIPS 2026Instruct-Particulate: Scaling Feed-Forward 3D Object Articulation with Kinematic Control
619. NeurIPS 2026InstructVVT: Instruction-Driven Video Virtual Try-On without Auxiliary Spatial Priors
620. NeurIPS 2026Integrating digital twins with randomized experiments
621. NeurIPS 2026Integrating Local and Global Entropy for Uncertainty Quantification in LLMs
622. NeurIPS 2026IntegrityBench: Can LLMs Be Trusted as Co-Scientists? A Research Integrity Benchmark
623. NeurIPS 2026Intend, Reflect, Refine: An Adaptive Multimodal Reflection Framework for Autonomous Driving
624. NeurIPS 2026IntentLens: Grounding Underspecified Multimodal Queries for Recommendation via Tool-Augmented Reasoning
625. NeurIPS 2026Interleaved Latent Thinking and Adaptive Termination for Efficient Reasoning LLMs
626. NeurIPS 2026Internalize External Competence for Visual Instruction Editing
627. NeurIPS 2026Intervention-Guided Image-Free Classifier Expansion for Fine-Grained Recognition
628. NeurIPS 2026Invertible Logits Transformation for Accuracy-Preserving Post-Hoc Uncertainty Calibration
629. NeurIPS 2026Iris: Empowering Video MLLMs with High-Frequency Pose Priors via Spatiotemporal Binding
630. NeurIPS 2026Is Text All You Need? Text as a Universal Information Bottleneck for Speech LLMs
631. NeurIPS 2026It Cancels: O(r²) Cholesky Updates for Whitened Operators
632. NeurIPS 2026It's All Training: A Fully Synthetic Single-Stage Recipe for LLMs
633. NeurIPS 2026Itô maps for any-step SDEs
634. NeurIPS 2026ITO: Multi-View Alignment and Training-Time Fusion for Image-Text Pretraining
635. NeurIPS 2026JailBound: A FOL-Guided Jailbreak Evaluation Framework for Revealing Safety Boundaries of LLMs
636. NeurIPS 2026JEDI: Real-Time Jailbreak Defense for LLMs via In-Generation Detection and Intervention
637. NeurIPS 2026JODA: Composable Joint Dynamics for Articulated Objects
638. NeurIPS 2026K12-KGraph: A Curriculum-Aligned Knowledge Graph for Benchmarking and Training Educational LLMs
639. NeurIPS 2026KINDER: Kernel-based Independence for Fair Representation Learning via Prototype-space Erasure
640. NeurIPS 2026Knowledge Localization in Mixture-of-Experts LLMs Using Cross-Lingual Inconsistency
641. NeurIPS 2026KV Packet: Recomputation-Free Context-Independent KV Caching for LLMs
642. NeurIPS 2026kVNN: Learnable Volterra Network Kernels
643. NeurIPS 2026LAION-BVD: A 10-Million-Hour Open Video Dataset for Multimodal Pre-training
644. NeurIPS 2026Lang-SVG: Hierarchical Image Vectorization with Language Priors
645. NeurIPS 2026LAPrune: Logits-Aligned Scoring Proxy for KV Pruning via Vector Quantization
646. NeurIPS 2026Large Discrete Policy: Advancing Explicit Behavior Modeling with Stochastic Iterative Scoring
647. NeurIPS 2026Large Language Model Selection with Limited Annotations
648. NeurIPS 2026Large Language Models Develop Belief State Geometry In-Context
649. NeurIPS 2026LaST-VLA: Thinking in Latent Spatio-Temporal Space for Vision-Language-Action in Autonomous Driving
650. NeurIPS 2026Latent Reasoning in Continuous Space for Unified Multimodal Models

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。