CVPR 2026 accepted papers on Vision-language and multimodal
869 of the 4,042 papers accepted at CVPR 2026 (21.5%) are on Vision-language and multimodal, compared with 15.6% at CVPR 2025. They fall into 6 topics.
See CVPR 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 · 438 papers
- Consensus Entropy: Harnessing Multi-VLM Agreement for Self-Verifying and Self-Improving OCR (1 citation)
- IBISAgent: Reinforcing Pixel-Level Visual Reasoning in MLLMs for Universal Biomedical Object Referring and Segmentation (1 citation)
- LLaDA-V: Large Language Diffusion Models with Visual Instruction Tuning (1 citation)
- OctoMed: Data Recipes for State-of-the-Art Multimodal Medical Reasoning (1 citation)
- UniChange: Unifying Change Detection with Multimodal Large Language Model (1 citation)
- 3D Space as a Scratchpad for Editable Text-to-Image Generation
- 4DP-QA: Scalable QA for 4D Perception in Vision Language Models
- 4DWorldBench: A Comprehensive Evaluation Framework for 3D/4D World Generation Models
- A More Word-like Image Tokenization for MLLMs
- Abstract 3D Perception for Spatial Intelligence in Vision-Language Models
- Act2See: Emergent Active Visual Perception for Video Reasoning
- AdaIAT: Adaptively Increasing Attention to Generated Text to Alleviate Hallucinations in LVLM
and 426 more in the interactive map.
Cross-modal learning · 138 papers
- Beyond Strict Pairing: Arbitrarily Paired Training for High-Performance Infrared and Visible Image Fusion (1 citation)
- Cross-Slice Knowledge Transfer via Masked Multi-Modal Heterogeneous Graph Contrastive Learning for Spatial Gene Expression Inference (1 citation)
- Customized Fusion: A Closed-Loop Dynamic Network for Adaptive Multi-Task-Aware Infrared-Visible Image Fusion (1 citation)
- Multi-Modal Image Fusion via Intervention-Stable Feature Learning (1 citation)
- RAMEN: Resolution-Adjustable Multimodal Encoder for Earth Observation (1 citation)
- A Mixed Diet Makes DINO An Omnivorous Vision Encoder
- Air-Know: Arbiter-Calibrated Knowledge-Internalizing Robust Network for Composed Image Retrieval
- AVION: Aerial Vision-Language Instruction from Offline Teacher to Prompt-Tuned Network
- BALM: A Model-Agnostic Framework for Balanced Multimodal Learning under Imbalanced Missing Rates
- Bidirectional Cross-Modal Prompting for Event-Frame Asymmetric Stereo
- Bilevel Layer-Positioning LoRA for Real Image Dehazing
- BioVITA: Biological Dataset, Model, and Benchmark for Visual-Textual-Acoustic Alignment
and 126 more in the interactive map.
Robot action and navigation · 100 papers
- SwiftVLA: Unlocking Spatiotemporal Dynamics for Lightweight VLA Models at Minimal Overhead (1 citation)
- ACoT-VLA: Action Chain-of-Thought for Vision-Language-Action Models
- ActiveVLA: Injecting Active Perception into Vision-Language-Action Models for Precise 3D Robotic Manipulation
- Adaptive Action Chunking at Inference-time for Vision-Language-Action Models
- Affordance Field Intervention: Enabling VLAs to Escape Memory Traps in Robotic Manipulation
- APEX: A Decoupled Memory-based Explorer for Asynchronous Aerial Object Goal Navigation
- AT-VLA: Adaptive Tactile Injection for Enhanced Feedback Reaction in Vision-Language-Action Models
- AtomicVLA: Unlocking the Potential of Atomic Skill Learning in Robots
- AVA-VLA: Improving Vision-Language-Action models with Active Visual Attention
- AwareVLN: Reasoning with Self-awareness for Vision-Language Navigation
- Beyond Success: Refining Elegant Robot Manipulation from Mixed-Quality Data via Just-in-Time Intervention
- Boosting Vision-Language-Action Finetuning with Feasible Action Neighborhood Prior
and 88 more in the interactive map.
CLIP and zero-shot · 92 papers
- fMRI-LM: Towards a Universal Foundation Model for Language-Aligned fMRI Understanding (1 citation)
- NAF: Zero-Shot Feature Upsampling via Neighborhood Attention Filtering (1 citation)
- A Closed-Form Solution for Debiasing Vision-Language Models with Utility Guarantees Across Modalities and Tasks
- AnomalyVFM -- Transforming Vision Foundation Models into Zero-Shot Anomaly Detectors
- Authorize-on-Demand: Dynamic Authorization with Legality-Aware Intellectual Property Protection for VLMs
- Beyond Heuristic Prompting: A Concept-Guided Bayesian Framework for Zero-Shot Image Recognition
- Bias Is a Subspace, Not a Coordinate: A Geometric Rethinking of Post-hoc Debiasing in Vision-Language Models
- Boosting Quantitive and Spatial Awareness for Zero-Shot Object Counting
- CAPT: Confusion-Aware Prompt Tuning for Reducing Vision-Language Misalignment
- CLIP Is Shortsighted: Paying Attention Beyond the First Sentence
- CLIP-like Model as a Foundational Density Ratio Estimator
- Cluster-Aware Neural Collapse Prompt Tuning for Long-Tailed Generalization of Vision-Language Models
and 80 more in the interactive map.
Human and embodied modeling · 74 papers
- MarkushGrapher-2: End-to-end Multimodal Recognition of Chemical Structures (1 citation)
- TDATR: Improving End-to-End Table Recognition via Table Detail-Aware Learning and Cell-Level Visual Alignment (1 citation)
- Agile Deliberation: Concept Deliberation for Subjective Visual Classification
- Beyond Semantic Search: Towards Referential Anchoring in Composed Image Retrieval
- Bridging Human Evaluation to Infrared and Visible Image Fusion
- CADFS: A Big CAD Program Dataset and Framework for Computer-Aided Design with Large Language Models
- CLAY: Conditional Visual Similarity Modulation in Vision-Language Embedding Space
- D2Dewarp: Dual Dimensions Geometric Representation Learning Based Document Image Dewarping
- Duala: Dual-Level Alignment of Subjects and Stimuli for Cross-Subject fMRI Decoding
- DuetSVG: Unified Multimodal SVG Generation with Internal Visual Guidance
- Fine-grained Image Aesthetic Assessment: Learning Discriminative Scores from Relative Ranks
- GeoSURGE: Geo-localization using Semantic Fusion with Hierarchy of Geographic Embeddings
and 62 more in the interactive map.
Visual question answering · 27 papers
- Act Like a Pathologist: Tissue-Aware Whole Slide Image Reasoning (1 citation)
- Beyond Multiple Choice: Verifiable OpenQA for Robust Vision-Language RFT
- BridgeEQA: Virtual Embodied Agents for Real Bridge Inspections
- CC-VQA: Conflict- and Correlation-Aware Method for Mitigating Knowledge Conflict in Knowledge-Based Visual Question Answering
- Deeper Thought, Weaker Aim: Understanding and Mitigating Perceptual Impairment during Reasoning in Multimodal Large Language Models
- Ego-Grounding for Personalized Question-Answering in Egocentric Videos
- Extending Embodied Question Answering from Perception to Decision
- Learning to Focus and Precise Cropping:A Reinforcement Learning Framework with Information Gaps and Grounding Loss for MLLMs
- M4-RAG: A Massive-Scale Multilingual Multi-Cultural Multimodal RAG
- MedFG-VQA: Low-Frequency Memory and Graph Attention for Lightweight Medical VQA
- MovieRecapsQA: A Multimodal Open-Ended Video Question-Answering Benchmark
- ReAG: Reasoning-Augmented Generation for Knowledge-based Visual Question Answering
and 15 more in the interactive map.