Multimodal LLMs papers, 2023–2026

Key findings

• Multimodal LLMs went from 1,226 papers in 2023 to 5,192 in 2026, roughly 4.2x in three years.
• NeurIPS publishes the most Multimodal LLMs work: 3,319 papers from 2023 to 2026, 30% of the four-venue total.
• The fastest-growing overlap is with Applications · computer vision: papers tagged with both rose from 35 in 2023 to 596 in 2026.
• Among the 4 topics under Trending Systems, it ranks #1 by 2026 paper count; the fastest grower in the group, AI Agents, is up 12x versus 4.2x here.

Representative papers

Spread across years and venues so every edition shows up

• 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

All Multimodal LLMs papers

1,074 papers, page 21 of 22

1001. NeurIPS 2025Vanish into Thin Air: Cross-prompt Universal Adversarial Attacks for SAM2
1002. NeurIPS 2025VCM: Vision Concept Modeling with Adaptive Vision Token Compression via Instruction Fine-Tuning
1003. NeurIPS 2025VESSA: Video-based objEct-centric Self-Supervised Adaptation for Visual Foundation Models
1004. NeurIPS 2025VIBE: Annotation-Free Video-to-Text Information Bottleneck Evaluation for TL;DR
1005. NeurIPS 2025ViCrit: A Verifiable Reinforcement Learning Proxy Task for Visual Perception in VLMs
1006. NeurIPS 2025Vid-SME: Membership Inference Attacks against Large Video Understanding Models
1007. NeurIPS 2025VidEmo: Affective-Tree Reasoning for Emotion-Centric Video Foundation Models
1008. NeurIPS 2025Video Perception Models for 3D Scene Synthesis
1009. NeurIPS 2025Video-R1: Reinforcing Video Reasoning in MLLMs
1010. NeurIPS 2025Video-SafetyBench: A Benchmark for Safety Evaluation of Video LVLMs
1011. NeurIPS 2025VideoCAD: A Dataset and Model for Learning Long‑Horizon 3D CAD UI Interactions from Video
1012. NeurIPS 2025VideoChat-R1.5: Visual Test-Time Scaling to Reinforce Multimodal Reasoning by Iterative Perception
1013. NeurIPS 2025VideoGameQA-Bench: Evaluating Vision-Language Models for Video Game Quality Assurance
1014. NeurIPS 2025VideoHallu: Evaluating and Mitigating Multi-modal Hallucinations on Synthetic Video Understanding
1015. NeurIPS 2025VideoLucy: Deep Memory Backtracking for Long Video Understanding
1016. NeurIPS 2025VideoRFT: Incentivizing Video Reasoning Capability in MLLMs via Reinforced Fine-Tuning
1017. NeurIPS 2025VideoVLA: Video Generators Can Be Generalizable Robot Manipulators
1018. NeurIPS 2025VIKI‑R: Coordinating Embodied Multi-Agent Cooperation via Reinforcement Learning
1019. NeurIPS 2025Virus Infection Attack on LLMs: Your Poisoning Can Spread "VIA" Synthetic Data
1020. NeurIPS 2025VisDiff: SDF-Guided Polygon Generation for Visibility Reconstruction, Characterization and Recognition
1021. NeurIPS 2025Vision as a Dialect: Unifying Visual Understanding and Generation via Text-Aligned Representations
1022. NeurIPS 2025Vision Foundation Models as Effective Visual Tokenizers for Autoregressive Generation
1023. NeurIPS 2025Vision Function Layer in Multimodal LLMs
1024. NeurIPS 2025Vision Transformers Don't Need Trained Registers
1025. NeurIPS 2025Vision Transformers with Self-Distilled Registers
1026. NeurIPS 2025Vision-centric Token Compression in Large Language Model
1027. NeurIPS 2025Vision‑Language‑Vision Auto‑Encoder: Scalable Knowledge Distillation from Diffusion Models
1028. NeurIPS 2025VisionThink: Smart and Efficient Vision Language Model via Reinforcement Learning
1029. NeurIPS 2025ViSpec: Accelerating Vision-Language Models with Vision-Aware Speculative Decoding
1030. NeurIPS 2025ViSPLA: Visual Iterative Self-Prompting for Language-Guided 3D Affordance Learning
1031. NeurIPS 2025Visual Diversity and Region-aware Prompt Learning for Zero-shot HOI Detection
1032. NeurIPS 2025Visual Instruction Bottleneck Tuning
1033. NeurIPS 2025Visual Structures Help Visual Reasoning: Addressing the Binding Problem in LVLMs
1034. NeurIPS 2025Visual Thoughts: A Unified Perspective of Understanding Multimodal Chain-of-Thought
1035. NeurIPS 2025VisualLens: Personalization through Task-Agnostic Visual History
1036. NeurIPS 2025VITA-1.5: Towards GPT-4o Level Real-Time Vision and Speech Interaction
1037. NeurIPS 2025VITRIX-CLIPIN: Enhancing Fine-Grained Visual Understanding in CLIP via Instruction-Editing Data and Long Captions
1038. NeurIPS 2025VITRIX-UniViTAR: Unified Vision Transformer with Native Resolution
1039. NeurIPS 2025VL-Rethinker: Incentivizing Self-Reflection of Vision-Language Models with Reinforcement Learning
1040. NeurIPS 2025VL-SAE: Interpreting and Enhancing Vision-Language Alignment with a Unified Concept Set
1041. NeurIPS 2025VLA-Cache: Efficient Vision-Language-Action Manipulation via Adaptive Token Caching
1042. NeurIPS 2025VLA-OS: Structuring and Dissecting Planning Representations and Paradigms in Vision-Language-Action Models
1043. NeurIPS 2025VLM in a flash: I/O-Efficient Sparsification of Vision-Language Model via Neuron Chunking
1044. NeurIPS 2025VLM-R³: Region Recognition, Reasoning, and Refinement for Enhanced Multimodal Chain-of-Thought
1045. NeurIPS 2025VLMLight: Safety-Critical Traffic Signal Control via Vision-Language Meta-Control and Dual-Branch Reasoning Architecture
1046. NeurIPS 2025VLMs can Aggregate Scattered Training Patches
1047. NeurIPS 2025VLMs have Tunnel Vision: Evaluating Nonlocal Visual Reasoning in Leading VLMs
1048. NeurIPS 2025VMDT: Decoding the Trustworthiness of Video Foundation Models
1049. NeurIPS 2025Vocabulary-Guided Gait Recognition
1050. NeurIPS 2025VQ-Seg: Vector-Quantized Token Perturbation for Semi-Supervised Medical Image Segmentation

FAQ

How many Multimodal LLMs papers were published at NeurIPS, ICLR, ICML and CVPR from 2023 to 2026?

11,209 in total: 1,226 in 2023, 1,716 in 2024, 3,075 in 2025, 5,192 in 2026.

How many Multimodal LLMs papers are at NeurIPS 2026?

NeurIPS 2026 has 1,366 Multimodal LLMs papers, versus 1,074 the year before.

How many Multimodal LLMs papers are at ICLR 2026?

ICLR 2026 has 1,401 Multimodal LLMs papers, versus 816 the year before.

How many Multimodal LLMs papers are at ICML 2026?

ICML 2026 has 1,051 Multimodal LLMs papers, versus 390 the year before.

How many Multimodal LLMs papers are at CVPR 2026?

CVPR 2026 has 1,374 Multimodal LLMs papers, versus 795 the year before.

Which conference publishes the most Multimodal LLMs papers?

NeurIPS, with 3,319 papers from 2023 to 2026.