Multimodal Medical Image Parsing with Task-Specific Low-Rank Adapters

cs.CV, cs.AI, cs.LG

Submitted: 2026-10-14

Updated: 2026-10-14

Comments: 17 pages, 4 figures. FLARE 2026 Task 3 methodology paper (2D track)

Code: https://github.com/TahaSabir0/FLARE26-Hexapus

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Sources

• CRIMSON: A Clinically-Grounded LLM-Based Metric for Generative Radiology Report Evaluation
• Qwen2.5-VL Technical Report
• Med-MoE: Mixture of Domain-Specific Experts for Lightweight Medical Vision-Language Models
• QwenVLConnector: A Fast, Unified Medical VLM Chatbot for Fine-Grained Clinical Perception and Text Generation
• Understanding Task Transfer in Vision-Language Models
• MedGemma Technical Report
• GLiClass: Generalist Lightweight Model for Sequence Classification Tasks
• Can Generalist Vision Language Models (VLMs) Rival Specialist Medical VLMs? Benchmarking and Strategic Insights

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