arXiv Computer Vision September 17, 2026 By Kiarash Tajbakhsh, Abdelrahman Faqieh, Michael Jopiti, Javier Garcia-Baroja, Philipp Zens, Branislav Zagrapan, Yuri Tolkach, Martin D. Berger, Aurel Perren, Bastian Dislich, Inti Zlobec, Amjad Khan

Lumen: Parameter-Efficient Alignment of Pretrained Vision and Language Encoders for Zero-Shot Computational Pathology

Read the original on arXiv Computer Vision →

Lumen is a pathology vision‑language model that aligns frozen unimodal foundation models (Virchow2 and BioMedBERT) using rank‑4 adapters and projection heads, training only 0.40% of the total parameters on the QUILT‑1M corpus. It achieves the highest mean chance‑corrected balanced accuracy (0.546) across nine zero‑shot patch benchmarks and demonstrates strong performance on lymph‑node metastasis detection, with AUROC scores of 0.964 internally and 0.955 externally. While it ranks third in cross‑modal retrieval, Lumen’s low‑parameter training yields competitive results at both patch and slide levels.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

arXiv Computer Vision

Aug 24

When Adaptation Hurts: Connecting Representational Drift to OOD Failures in MedSAM Fine-Tuning

arXiv:2608.21300v1 Announce Type: new Abstract: Foundation models for medical image segmentation, like prompt-based MedSAM, generalize well across domains and modalities, often in zero or few-shot se...

By Marko Haralovi\'c, Sounic Akkaraju, Carlo Baretta, Vasil Zapryanov, Alexia Briassouli

computer-vision fine-tuning benchmarks safety

arXiv AI

Aug 11

Simple Token-Efficient Vision-Language Model for Case-level Pathology Synoptic Report Generation

arXiv:2605. 30716v2 Announce Type: replace-cross Abstract: Generating clinically useful pathology reports for pathology cases from whole-slide images (WSIs) is challenging due to gigapixel resolution, long visual-token sequences, and the complexity of case-level reasoning, where a single case may contain multiple WSIs with heterogeneous tissues and ambiguous findings.

By Zhiyuan Yang, Jiahao Cheng, Vincent Quoc-Huy Trinh, Mahdi S. Hosseini

llms fine-tuning multimodal

arXiv AI

Aug 7

Big, Bright, or Invisible: A Frozen-Feature Benchmark of 3D CT Foundation Models

arXiv:2608. 05960v1 Announce Type: cross Abstract: Routine CT interpretation is inherently comprehensive, capturing incidental findings across the entire scan volume.

By Maulik Chevli, Johannes Brandt, Rickmer Braren, Daniel Rueckert, Philip M\"uller

llms rag nlp multimodal benchmarks safety

arXiv AI

Jun 24

HiPath: Hierarchical Vision-Language Alignment for Structured Pathology Report Prediction

arXiv:2603. 19957v2 Announce Type: replace-cross Abstract: Pathology reports are structured, multi-granular documents encoding diagnostic conclusions, histological grades, and ancillary test results across one or more anatomical sites; yet existing pathology vision-language models (VLMs) reduce this output to a flat label or free-form text.

By Ruicheng Yuan, Zhenxuan Zhang, Anbang Wang, Liwei Hu, Xiangqian Hua, Yaya Peng, Jiawei Luo, Guang Yang

llms multimodal safety

arXiv Computer Vision

Aug 25

LanGuSTE: Language-Guided Coarse-to-Fine Patch Selection for Efficient Whole Slide Image Analysis

LanGuSTE is a patch‑selection framework for whole slide image analysis that uses vision‑language models and large language model knowledge. It introduces Cross‑Scale Visual Prompt Tuning to align low‑resolution and high‑resolution patches, and a coarse‑to‑fine selection module that encodes only informative high‑resolution patches. Experiments show LanGuSTE cuts overall processing time to about one‑third of the baseline while matching or surpassing diagnostic performance of exhaustive and state‑of‑the‑art methods.

By Yonghan Shin, Gangsu Kim, Won-Ki Jeong

llms multimodal benchmarks

Hugging Face Trending Papers

Jul 23

Do Pathology Vision-Language Models Truly See Pathology?

Pathology vision-language models (VLMs) have recently progressed rapidly and are commonly evaluated by answer accuracy on pathology VQA benchmarks. However, we dig into current evaluations and identify three overlooked issues: 1) Visual evidence is not always necessary.

llms multimodal benchmarks

One email a morning, machine-written

Related stories

When Adaptation Hurts: Connecting Representational Drift to OOD Failures in MedSAM Fine-Tuning

Simple Token-Efficient Vision-Language Model for Case-level Pathology Synoptic Report Generation

Big, Bright, or Invisible: A Frozen-Feature Benchmark of 3D CT Foundation Models

HiPath: Hierarchical Vision-Language Alignment for Structured Pathology Report Prediction

LanGuSTE: Language-Guided Coarse-to-Fine Patch Selection for Efficient Whole Slide Image Analysis

Do Pathology Vision-Language Models Truly See Pathology?