Cvpr 2026 Findings 2026

761 papers

Top Research Directions

Research directions most represented by Cvpr 2026 Findings 2026 papers.

The Decoupling of Visual Perception and Logical Reasoning 10
The Spatial Reasoning Bottleneck in Multimodal Models 9
Generalizing from Synthetic Data to the Real World 9
Scaling Visual Attention Beyond Quadratic Complexity 9
The Cross Modal Safety Gap in Multimodal Models 7
Scaling Spatiotemporal Attention Beyond Quadratic Complexity 7
Group Relative Policy Optimization for Large Reasoning Models 6
Spatial Attention Masking in Generative Diffusion Models 6
Modality Imbalance in Multimodal Large Language Models 6
Managing Visual Token Redundancy in Multimodal Models 6
The Representational Bottleneck of Frozen Foundation Models 5
Overcoming Low Frequency Bias with Spectral Losses 5
Closed Loop Visual Feedback for Multimodal Generation 5
Aligning Video Generation Models with Physical Laws 5
Zero-Shot Generalization through Diverse Synthetic Supervision 5

Top Authors

AuthorPapers
Andrew Feng1
Peter Anthony Beerel1
Zhenglong Sun1

Papers

Title
MeMix: Multi-Encoder Mixture Framework for Medical Report Generation
Memorization in 3D Shape Generation: An Empirical Study
Memorization In Stable Diffusion Is Unexpectedly Driven by CLIP Embeddings
Memory-Efficient Continual Learning with Prototypical Exemplar Condensation
MHMamba: Multi-Head Mamba for 3D Brain Tumor Segmentation
Mining Real-World Image Relations for Large-Scale Controllable Generation and Editing
MipKV: A Sparsify-then-Recover Paradigm for Accelerating Large Vision-Language Model Pre-Filling
MIRA: Multimodal Iterative Reasoning Agent for Image Editing
Mitigating Batch Effects in Histopathology via Language-Mediated Robust Embedding Generation
Mitigating Information Forgetting via Entropy-Driven Progressive Retrospection for Multimodal Long Reasoning
Mitigating Object Hallucinations in LVLMs via Attention Imbalance Rectification
Mitigating the ID–OOD Tradeoff in Open-Set Test-Time Adaptation
Mitigating Vision-Text Order Bias in Vision-Language Model
Mitigating Visual Context Degradation in Large Multimodal Models: A Training-Free Decoupled Agentic Framework
MiVLA: Towards Generalizable Vision-Language-Action Model with Human-Robot Mutual Imitation Pre-Training
Mix-to-Max: Optimizing Data Mixtures for Peak Vision-Language Efficiency
mmDiff: A Noise-Robust Differentiable Ray-Tracing Framework for mmWave Scene Calibration and Channel Prediction
MMR1: Enhancing Multimodal Reasoning with Variance-Aware Sampling
Modality-Aware and Anatomical Vector-Quantized Autoencoding for Multimodal Brain MRI
Modality-Aware Bit Allocation for Mixed-Precision Quantization of Vision-Language Models
Model Merging on Loss Landscapes: A Geometric Perspective
Modulate-and-Map: Crossmodal Feature Mapping with Cross-View Modulation for 3D Anomaly Detection
MolRecBench-Wild: A Real-World Benchmark for Optical Chemical Structure Recognition
MoonSeg3R: Monocular Online Zero-Shot Segment Anything in 3D with Reconstructive Foundation Priors
MOSSTrack : Modality-Specific Spatio-Temporal Context Learning for RGB-T Tracking
MotionDuet: Dual-Conditioned 3D Human Motion Generation with Video-Regularized Text Learning
MoVieDrive: Urban Scene Synthesis with Multi-Modal Multi-View Video Diffusion Transformer
MPerS: Dynamic MLLM MixExperts Perception-Guided Remote Sensing Scene Segmentation
M-PhyGs: Multi-Material Object Dynamics from Video
MReactor: Offline Multiple Appropriate Facial Reaction Generation with Hierarchical Cognitive Disentanglement
Mull-Tokens: Modality-Agnostic Latent Thinking
Multimodal Reasoning with Explicit Reasoning Patterns and Rewards
MuSCM: Mutual Spatial Correlation Mapping for Class Incremental Detection Transformer
MVSSM: Motion-Aware Visual State Space Model for Efficient Video Deblurring
Myopia Rectification: KV Cache Pruning for MLLMs Via Dynamic Attention Subsidy and Token Reclamation
NAKUL-Med: Spectral-Graph State Space Models with Dynamics Kernels for Medical Signals
Name That Part: 3D Part Segmentation and Naming
NCSTR: Node-Centric Decoupled Spatio-Temporal Reasoning for Video-Based Human Pose Estimation
NeVStereo: A NeRF-Driven NVS-Stereo Architecture for High-Fidelity 3D Tasks
No Cache Left Idle: Accelerating Diffusion Model via Extreme-Slimming Caching
Now You See It, Now You Don't: Instant Concept Erasure for Safe Text-to-Image and Video Generation
NRFP: A Noise-Robust Feature Plugin for Source-Free Domain Adaptation
NSGuard: Null-Space Guided Robust Watermarking for Data Copyright Protection in Customized Generation
Object-Centric Vision Token Pruning for Vision Language Models
Object Pose Transformer: Unifying Unseen Object Pose Estimation
Objects in Generated Videos Are Slower Than They Appear: Models Suffer Sub-Earth Gravity and Don’t Know Galileo’s Principle...for now
ODOV: Benchmark the Open-Domain Open-Vocabulary Object Detection
OffNadirLoc: Benchmark and Framework for Challenging UAV-to-Satellite Geo-Localization Under Large Off-Nadir Views
OKGraph: Online Knowledge Graph Probing for Open-Vocabulary Recognition
OminiControl2: Efficient Conditioning for Diffusion Transformers

« prev Page 9/16 next »