NeurIPS 2026 accepted papers on Vision-language and multimodal
536 of the 6,080 papers accepted at NeurIPS 2026 (8.8%) are on Vision-language and multimodal, compared with 9.3% at NeurIPS 2025. They fall into 6 topics.
See NeurIPS 2026 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Multimodal LLMs · 253 papers
- 3D-PLOT-LLM: Part-Level Object Tokens for 3D Large Language Models
- A Multimodal Benchmark for Evaluating Cause-of-Death Inference Using Child Health and Mortality Data
- A systematic evaluation of vision-language models for observational astronomical reasoning tasks
- AgroOmni: A Large-Scale Multi-view Agricultural Dataset for Cross-Scale Multimodal Reasoning
- Allocentric Perceiver: Disentangling Allocentric Reasoning from Egocentric Visual Priors via Frame Instantiation
- AlloSpatial: Agentic Harness Framework for Spatial Reasoning in Foundation Models
- ARK: A Dual-Axis Multimodal Retrieval Benchmark along Reasoning and Knowledge
- ATLAS: Agentic or Latent Visual Reasoning? One Word is Enough for Both
- BalCapRL : A Balanced Framework for RL-Based MLLM Image Captioning
- Benchmarking Multimodal Mathematical Reasoning with Explicit Visual Dependency
- Bernini: Latent Semantic Planning for Video Diffusion
- Beyond Bag-of-Words: Diagnosing Compositional Binding Failures in Vision-Language Models
and 241 more in the interactive map.
Cross-modal learning · 94 papers
- EchoXFlow: A Beamspace Echocardiography Dataset for Cardiac Motion, Flow, and Function (3 citations)
- MOSAIC-CONUS: A Multimodal, Multi-Temporally Paired dataset for Earth Sciences (1 citation)
- Asymmetric Hierarchical Anchoring for Robust Audio–Visual Cross-Modal Generalization
- BayesRAG: Probabilistic Mutual Evidence Corroboration for Multimodal Retrieval-Augmented Generation
- Beyond Modality Fusion: Deep Ensembles for Multimodal Classification
- Beyond Spatial-Domain Supervision: A Relation Constrained Space for Multi-Modal Image Fusion
- Brain-OF: An Omnifunctional Foundation Model for fMRI, EEG and MEG
- Breaking Modality Heterogeneity in Low-Bit Quantization for Large Vision-Language Models
- Closing the Indexing-Decoding Gap in Multimodal Generative Retrieval via Prefix Retention Optimization
- CORTEG: Foundation Models Enable Cross-Modality Representation Transfer from Scalp to Intracranial Brain Recordings
- Do More Modalities Always Help? A Geometric Perspective on Missing-Modality Robustness
- Drift-Aware Multimodal User Representation Learning via Multi-Scale Temporal Modeling and Sparse Mixture-of-Experts
and 82 more in the interactive map.
Robot action and navigation · 90 papers
- ActionUNet: Improving Robustness of VLA Models with Efficient Multi-scale Fine-tuning
- APIVOT: Adaptive Planning with Interleaved Vision-Language Thoughts
- ATI-VLA: Action-Centric Predictive Vision–Language–Action Models via Actionable Alignment Then Adaptive Injection
- Beyond Appearance Shifts: Task-Semantic Action Calibration for VLA Models
- Beyond World-Frame Action Heads: Motion-Centric Action Frames for Vision-Language-Action Models
- ChainFlow-VLA: Causal Flow Planning with Vision-Language Models
- Code2World: A GUI World Model via Renderable Code Generation
- CofactVLA: Deconfounding Vision-Language-Action Models via Counterfactual Intervention
- DriveDreamer-Policy: A Geometry-Grounded World-Action Model for Unified Generation and Planning
- DynaFLIP: Rethinking Robotics Perception via Tri-Modal-Dynamics Guided Representation
- ECHO: Continuous Hierarchical Memory for Vision-Language-Action Models
- ElegantVLA: Learning When to Think for Efficient Vision-Language-Action Models
and 78 more in the interactive map.
Human and embodied modeling · 58 papers
- Attention Alignment Between Humans and Vision-Language Models (1 citation)
- A Systematic Evaluation of Co-folding Model Representations for Small-Molecule Learning
- A World Model of Radiologist Reading for Medical Image Representation Learning
- Attributions All the Way Down? The Metagame of Interpretability
- Characterizing Universal Object Representations Across Vision Models
- Concepts Worth Having: Refining VLM-Guided Concept Bottleneck Models with Minimal Annotations
- Do Composed Image Retrieval Benchmarks Require Multimodal Composition?
- Half-Truths Break Similarity-Based Retrieval
- Language-Assisted Image Clustering Guided by Discriminative Relational Signals and Adaptive Semantic Centers
- LINE: LLM-based Iterative Neuron Explanations for Vision Models
- M*: A Modular, Extensible, Serving System for Multimodal Models
- MemLeak: Diagnosing Information Leaks in Multimodal Agent Memory
and 46 more in the interactive map.
CLIP and zero-shot · 29 papers
- Align3D-AD: Cross-Modal Feature Alignment and Dual-Prompt Learning for Zero-shot 3D Anomaly Detection
- Aligning Forest and Trees in Images & Long Captions for Visually Grounded Understanding
- AmaraSpatial-10K: A Spatially and Semantically Aligned 3D Dataset for Spatial Computing and Embodied AI
- Beyond Interpretability: When, Why, and How Sparse Autoencoders Enable Label-Free Visual Steering
- C3P: Contrastive promoter-protein pretraining yields representations capturing bacterial gene regulation
- Cracks in the Foundation: A Civil Infrastructure Dataset to Challenge Vision Foundation Models
- MindAlign: Bridging EEG, Vision, and Language for Zero-Shot Visual Decoding
- PriorVLA: Prior-Preserving Adaptation for Vision-Language-Action Models
- ReefNet: A Large-Scale Dataset and Benchmark for Fine-Grained Coral Reef Recognition
- Reinforcing VLAs in Task-Agnostic World Models
- Text as Partial Constraint: Core–Residual Alignment for Robust Vision–Language Learning
- What Drives Test-Time Adaptation for CLIP? A Controlled Empirical Study from an Update Perspective
and 17 more in the interactive map.
Visual question answering · 12 papers
- CoMet: Context and Multiplicity Decomposition for Multimodal Uncertainty Estimation
- Exemplar2VQA: A Scalable Exemplar-Driven Visual Question Answering Generation Framework via Multi-Agent Coding
- MedVIGIL: Evaluating Trustworthy Medical VLMs Under Broken Visual Evidence
- SIEVES: Selective Prediction Generalizes through Visual Evidence Scoring
- The World is Not Mono: Enabling Spatial Understanding in Large Audio-Language Models
- EDMA: Entropy-Driven Multimodal Answering
- Not Another Text Benchmark: Putting the “Visual" Back in Visual Question Answering for Large Video Models
- OmniSE: Support-Faithful Optimization for Evidence-Centric Audio-Video Reasoning
- RSRCC: A Remote Sensing Regional Change Comprehension Benchmark Constructed via Retrieval-Augmented Best-of-𝑁 Ranking
- Selective Answering for Medical VQA via Parallel Independent Claim Verification
- SitCom: Scaling Egocentric Multi-Party Spoken Dialogue for Situated Communication Assistance
- Trust What Matters: Language-Conditioned Evidence Routing for Video-IMU Action Question Answering