analiza aiqbench
This research addresses a critical gap in medical AI by introducing the first cross-scale reasoning paradigm for vision-language models (VLMs) applied to pathology. The authors present Scale-VQA, a new benchmark designed to prevent text-only shortcuts and force models to integrate evidence across different tissue magnifications using reinforcement learning. This work advances the field of specialized computer vision by demonstrating that explicit multi-magnification supervision significantly improves pathological understanding beyond single-scale benchmarks.
Enhancing Pathological VLMs with Cross-scale Reasoning
arXiv:2606.17412v4 Announce Type: replace-cross Abstract: Pathological images are inherently multi-scale, requiring pathologists to integrate evidence from global tissue architecture at low magnification to cellular morphology at higher magnification for accurate diagnosis. While existing pathological datasets for vision-language models (VLMs) include various scales, they often lack explicit cross-scale reasoning objectives. This limitation prevents VLMs from capturing essential cross-scale representations and learning evidence-based reasoning. To bridge this gap, we introduce the first cross-scale training and evaluation paradigm that formulates pathology interpretation as multi-magnification reasoning. However, creating such a task reveals a critical challenge: multi-image visual question answering (VQA) is prone to text-only shortcuts, which allow models to guess answers using magnification-dependent artifacts rather than visual evidence. To address this, we propose a leakage-aware curation pipeline that combines adversarial text-only screening with constraint-guided question design. Using this pipeline, we construct Scale-VQA, a high-quality benchmark with 4,685 multiple-choice questions grounded in 2,537 pathology images across multiple magnification levels. Finally, we present ScaleReasoner-R1, a model trained via reinforcement learning to optimize performance on cross-scale VQA tasks. ScaleReasoner-R1 achieves state-of-the-art performance on our cross-scale reasoning benchmark and generalizes to SOTA performance on established single-scale benchmarks. Findings suggest that even the limited cross-scale supervision can significantly improve pathological understanding. Code is available at https://github.com/iMVR-PL/ScaleReasoner-R1.
Abstract:Pathological images are inherently multi-scale, requiring pathologists to integrate evidence from global tissue architecture at low magnification to cellular morphology at higher magnification for accurate diagnosis. While existing pathological datasets for vision-language models (VLMs) include various scales, they often lack explicit cross-scale reasoning objectives. This limitation prevents VLMs from capturing essential cross-scale representations and learning evidence-based reasoning. To bridge this gap, we introduce the first cross-scale training and evaluation paradigm that formulates pathology interpretation as multi-magnification reasoning. However, creating such a task reveals a critical challenge: multi-image visual question answering (VQA) is prone to text-only shortcuts, which allow models to guess answers using magnification-dependent artifacts rather than visual evidence. To address this, we propose a leakage-aware curation pipeline that combines adversarial text-only screening with constraint-guided question design. Using this pipeline, we construct Scale-VQA, a high-quality benchmark with 4,685 multiple-choice questions grounded in 2,537 pathology images across multiple magnification levels. Finally, we present ScaleReasoner-R1, a model trained via reinforcement learning to optimize performance on cross-scale VQA tasks. ScaleReasoner-R1 achieves state-of-the-art performance on our cross-scale reasoning benchmark and generalizes to SOTA performance on established single-scale benchmarks. Findings suggest that even the limited cross-scale supervision can significantly improve pathological understanding. Code is available at this https URL.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
arXiv:2606.17412 [cs.CV]
(or arXiv:2606.17412v4 [cs.CV] for this version)
Text integral sub licența deschisă a publicației · 221 cuvinte
Subiecte conexe
Enhancing Pathological VLMs with Cross-scale Reasoning
arXiv · 17 iun. 2026
PathScale-R1: Cross-scale Reasoning for Pathological Image Analysis
arXiv · 28 iul. 2026
Amazon is investing in the Lean Focused Research Organization
Amazon Science · 26 iul. 2026
34 Amazon Research Awards Build on Trainium recipients announced
Amazon Science · 5 aug. 2026