CARVE: Cross-Slice Anisotropic Reallocation of Visual Evidence for Efficient 3D Medical Volume Understanding
Zhenyu Yi, Qiang Hu, Zhenhao Li, Jiaxuan Zhao, Yusong Sun, Lichi Zhang
cs.CV, cs.AI, cs.CL
Submitted: 2026-08-05
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- M3D: Advancing 3D Medical Image Analysis with Multi-Modal Large Language Models
- Token Merging: Your ViT But Faster
- MMTok: Multimodal Coverage Maximization for Efficient Inference of VLMs
- Photon: Speedup Volume Understanding with Efficient Multimodal Large Language Models
- FoPru: Focal Pruning for Efficient Large Vision-Language Models
- Hulu-Med: A Transparent Generalist Model towards Holistic Medical Vision-Language Understanding
- OmniV-Med: Scaling Medical Vision-Language Model for Universal Visual Understanding
- Geometry-Guided 3D Visual Token Pruning for Video-Language Models
- E-MRL: Cross-view Aligned Evidence-driven Multimodal Reinforcement Learning for Reliable 3D Tumor Analysis
- SeGPruner: Semantic-Geometric Visual Token Pruner for 3D Question Answering
- Not All Patches are What You Need: Expediting Vision Transformers via Token Reorganizations
- MedPruner: Training-Free Hierarchical Token Pruning for Efficient 3D Medical Image Understanding in Vision-Language Models
- MMG-Vid: Maximizing Marginal Gains at Segment-level and Token-level for Efficient Video LLMs
- MedGemma Technical Report
- LLaVA-PruMerge: Adaptive Token Reduction for Efficient Large Multimodal Models
- Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
- Towards Generalist Foundation Model for Radiology by Leveraging Web-scale 2D&3D Medical Data
- Brain-Adapter: A Dual-Stream Vision-Language MIL Framework for Comprehensive 3D CT Diagnosis of Acute Intracranial Pathologies
- MedFrameQA: A Multi-Image Medical VQA Benchmark for Clinical Reasoning
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models