TPAMI 2026 papers on Vision-language and multimodal
105 of the 880 papers published in TPAMI 2026 (11.9%) are on Vision-language and multimodal, compared with 7.7% at TPAMI 2025. They fall into 6 topics.
See TPAMI 2026 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Multimodal LLMs · 37 papers
- Lingshu: Generalist Foundation Model for Unified Multimodal Medical Understanding and Reasoning (319 citations)
- TextMonkey: An OCR-Free Large Multimodal Model for Understanding Document (243 citations)
- Beyond LLaVA-HD: Diving Into High-Resolution Multimodal Large Language Models (83 citations)
- MERBench: A Unified Evaluation Benchmark for Multimodal Emotion Recognition (71 citations)
- StructChart: On the Schema, Metric, and Augmentation for Visual Chart Understanding (64 citations)
- VLBiasBench: A Comprehensive Benchmark for Evaluating Bias in Large Vision-Language Model (51 citations)
- AeroVerse: UAV-Agent Benchmark Suite for Simulating, Pre-Training, Finetuning, and Evaluating Aerospace Embodied Foundation Models (39 citations)
- OmniParser V2: Structured-Points-of-Thought for Unified Visual Text Parsing and Its Generality to Multimodal Large Language Models (38 citations)
- Evaluating and Mitigating Relationship Hallucinations in Large Vision-Language Models (32 citations)
- AtomThink: Multimodal Slow Thinking With Atomic Step Reasoning (28 citations)
- RingMo-Agent: A Unified Remote Sensing Foundation Model for Multi-Platform and Multi-Modal Reasoning (28 citations)
- Simulating the Real World: A Unified Survey of Multimodal Generative Models (24 citations)
and 25 more in the interactive map.
Cross-modal learning · 31 papers
- A Comprehensive Survey on Multimodal Recommender Systems: Taxonomy, Evaluation, and Future Directions (167 citations)
- Dictionary Multi-Modal Temporal Graph Learning (11 citations)
- Decoupled Hierarchical Distillation for Multimodal Emotion Recognition (5 citations)
- Multiple Modalities Image Matching With Large-Scale Pre-Training (4 citations)
- Deep Information-Balanced Multimodal Learning (3 citations)
- An Efficient Regenerated Cross-Modal Hashing: Improving Existing Hash Codes With the Arbitrary Length (2 citations)
- Diff-MM: Exploring Pre-Trained Text-to-Image Generation Models for Unified Multi-Modal Object Tracking (2 citations)
- Generalizable Egocentric Task Verification via Cross-Modal Hybrid Hypergraph Matching (2 citations)
- Alignment-Consistent Multimodal Learning under Uncertain Correspondence (1 citation)
- HIMO: Cross-Arbitrary-Modality Image Invariant Feature Transform With Hierarchical Intrinsic Major Orientation (1 citation)
- How Far are We From Generating Missing Modalities With Foundation Models? (1 citation)
- Incomplete Multimodal Probability Flow Recovery for Emotion Recognition (1 citation)
and 19 more in the interactive map.
Human and embodied modeling · 13 papers
- Grad-ECLIP: Gradient-Based Visual and Textual Explanations for CLIP (14 citations)
- Concept Drift and Long-Tailed Distribution in Fine-Grained Visual Categorization: Benchmark and Method (10 citations)
- Distribution-to-Points Matching for Image Text Retrieval (3 citations)
- ChatTracker: Enhancing Visual Tracking via LLM-Driven Iterative Description Refinement (2 citations)
- Confidence-Aware Pseudo-Label Self-Correction for Weakly Supervised Visual Grounding (2 citations)
- Fine-Grained Alignment Supervision Matters in Vision-and-Language Navigation (2 citations)
- Codebook Transfer With Vision-to-Language Translation for Vector Quantization (1 citation)
- Bridging the Gap in Exam Handwritten Text Recognition: Dataset, Benchmark, and Modeling
- Discovering Visual Co-occurrence for Fine-grained Image Retrieval
- Generative Retrieval for Unsupervised Text-Based Person Search
- Learning Long-Form Movie Prior via Large Language Models
- Unified Multi-Subject Image Reconstruction From Human Brain Activity
and 1 more in the interactive map.
CLIP and zero-shot · 11 papers
- Text4Seg++: Advancing Image Segmentation via Generative Language Modeling (9 citations)
- ZUMA: Training-Free Zero-Shot Unified Multimodal Anomaly Detection (8 citations)
- Negation-Aware Test-Time Adaptation for Vision-Language Models (5 citations)
- DynaPURLS: Dynamic Refinement of Part-Aware Representations for Skeleton-Based Zero-Shot Action Recognition (2 citations)
- SDPT: Synchronous Dual Prompt Tuning for Visual-Language Pre-Trained Models (2 citations)
- RAW-Adapter: Adapting Pre-Trained Visual Model to Camera RAW Images and a Benchmark (1 citation)
- Advancing In-Context Learning for Efficient and Stable Medical Report Generation
- Difference-Aware Multi-Grained Feature Learning With Memory-Attended Joint Interaction Transformer for Image Difference Captioning
- HyperCLIP++: Fine-Tuning CLIP for Open-Vocabulary Semantic Segmentation in Hyperbolic Space
- MMA++: Effective Multi-Modal Adaptation for Vision-Language Models
- MuSc-V2: Zero-Shot Multimodal Industrial Anomaly Classification and Segmentation With Mutual Scoring of Unlabeled Samples
Robot action and navigation · 9 papers
- Dream to Recall: Imagination-Guided Experience Retrieval for Memory-Persistent Vision-and-Language Navigation (5 citations)
- Goal-Guided Prompting With Adaptive Modality Selection for Efficient Assembly Activity Anticipation in Egocentric Videos (3 citations)
- EgoMotion: Hierarchical Vision-Language Learning and Diffusion for Egocentric Motion Generation (2 citations)
- EgoLoc: A Generalizable Solution for Temporal Interaction Localization in Egocentric Videos (1 citation)
- YOTO++: Learning Long-Horizon Closed-Loop Bimanual Manipulation From One-Shot Human Video Demonstrations (1 citation)
- Integrating Agglomerative Perception With One-Step Action Generation for Robotic Manipulation
- MindExplore+: Towards Long-Horizon Vision-Language-Action System
- Toward a Causally-Inspired Evolving World Model for Vision-and-Language Navigation in Continuous Environments
- A Survey on Efficient Vision-Language-Action Models
Visual question answering · 4 papers
- Q-SiT: Teaching LMMs for Image Quality Scoring and Interpreting (21 citations)
- Parse, Align and Aggregate: Graph-Driven Compositional Reasoning for Video Question Answering (1 citation)
- Robust Audio-Visual Question Answering With Missing Modality in Training and Testing
- SHC: Deeply Activating Human-Like Cognitive Ability for Visual Question Answering
