{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,21]],"date-time":"2026-10-21T16:20:10Z","timestamp":1792599610035,"version":"4.5.0"},"reference-count":46,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,5,15]],"date-time":"2026-05-15T00:00:00Z","timestamp":1778803200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>\n                    Foundation vision-language models (VLMs) excel on natural images, but their utility for biomedical microscopy remains underexplored. In this paper, we investigate how in-context learning enables state-of-the-art VLMs to perform few-shot object detection when large annotated datasets are unavailable, as is often the case with microscopic images. We introduce the Micro-OD benchmark, a curated collection of 252 images specifically curated for in-context learning, with bounding-box annotations spanning 11 cell types across four sources, including two in-lab expert-annotated sets. We systematically evaluate eight VLMs under few-shot conditions and compare variants with and without implicit test-time reasoning tokens. We further implement a hybrid Few-Shot Object Detection (FSOD) pipeline that combines a detection head with a VLM-based few-shot classifier, which enhances the few-shot performance of recent VLMs on our benchmark. Across datasets, we observe that zero-shot performance is weak due to the domain gap; however, few-shot support consistently improves detection, with marginal gains achieved after six shots. We observe that some reasoning variant models show task-specific gains, but the effect varies across models and settings. Our results highlight in-context adaptation as a promising research direction requiring further development for microscopy, and our benchmark provides a reproducible testbed for advancing open-vocabulary detection in biomedical imaging. Our project page is at:\n                    <jats:ext-link>here<\/jats:ext-link>\n                    .\n                  <\/jats:p>","DOI":"10.3389\/frai.2026.1761903","type":"journal-article","created":{"date-parts":[[2026,5,15]],"date-time":"2026-05-15T05:44:16Z","timestamp":1778823856000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["In-context adaptation of VLMs for few-shot cell detection in optical microscopy"],"prefix":"10.3389","volume":"9","author":[{"given":"Shreyan","family":"Ganguly","sequence":"first","affiliation":[{"name":"Iowa State University","place":["Ames, IA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Angona","family":"Biswas","sequence":"additional","affiliation":[{"name":"Iowa State University","place":["Ames, IA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jaydeep","family":"Rade","sequence":"additional","affiliation":[{"name":"Iowa State University","place":["Ames, IA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Md Hasibul Hasan","family":"Hasib","sequence":"additional","affiliation":[{"name":"Iowa State University","place":["Ames, IA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nabila","family":"Masud","sequence":"additional","affiliation":[{"name":"Iowa State University","place":["Ames, IA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nitish","family":"Singla","sequence":"additional","affiliation":[{"name":"Iowa State University","place":["Ames, IA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Abhipsa","family":"Dash","sequence":"additional","affiliation":[{"name":"Iowa State University","place":["Ames, IA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ushashi","family":"Bhattacharjee","sequence":"additional","affiliation":[{"name":"Iowa State University","place":["Ames, IA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Soumik","family":"Sarkar","sequence":"additional","affiliation":[{"name":"Iowa State University","place":["Ames, IA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aditya","family":"Balu","sequence":"additional","affiliation":[{"name":"Iowa State University","place":["Ames, IA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anwesha","family":"Sarkar","sequence":"additional","affiliation":[{"name":"Iowa State University","place":["Ames, IA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Adarsh","family":"Krishnamurthy","sequence":"additional","affiliation":[{"name":"Iowa State University","place":["Ames, IA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,5,15]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2303.08774","article-title":"GPT-4 technical report","author":"Achiam","year":"2023","journal-title":"arXiv"},{"key":"B2","first-page":"4767","article-title":"\u201cFew-shot image classification benchmarks are too far from reality: build back better with semantic task sampling,\u201d","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops","author":"Bennequin","year":"2022"},{"key":"B3","doi-asserted-by":"publisher","first-page":"849","DOI":"10.1038\/nmeth.4397","article-title":"Data-analysis strategies for image-based cell profiling","volume":"14","author":"Caicedo","year":"2017","journal-title":"Nat. Methods"},{"key":"B4","first-page":"213","article-title":"\u201cEnd-to-end object detection with transformers,\u201d","author":"Carion","year":"2020","journal-title":"Computer Vision"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2507.06261","article-title":"Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities","author":"Comanici","year":"2025","journal-title":"arXiv"},{"key":"B6","article-title":"An image is worth 16x16 words : transformers for image recognition at scale","author":"Dosovitskiy","year":"2021","journal-title":"arXiv"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2411.13317","article-title":"Teaching VLMs to localize specific objects from In-context examples","author":"Doveh","year":"2025","journal-title":"arXiv"},{"key":"B8","doi-asserted-by":"publisher","first-page":"1038","DOI":"10.1038\/s41592-021-01249-6","article-title":"Livecell\u2013a large-scale dataset for label-free live cell segmentation","volume":"18","author":"Edlund","year":"2021","journal-title":"Nat. Methods"},{"key":"B9","doi-asserted-by":"crossref","first-page":"1411","DOI":"10.1109\/APSIPAASC47483.2019.9023120","article-title":"\u201cDistilling knowledge for non-neural networks,\u201d","volume-title":"2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","author":"Fukui","year":"2019"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2104.13921","article-title":"Open-vocabulary object detection via vision and language knowledge distillation","author":"Gu","year":"2022","journal-title":"arXiv"},{"key":"B11","doi-asserted-by":"crossref","first-page":"5351","DOI":"10.1109\/CVPR.2019.00550","article-title":"\u201cLVIS: a dataset for large vocabulary instance segmentation,\u201d","volume-title":"2019 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Gupta","year":"2019"},{"key":"B12","doi-asserted-by":"publisher","first-page":"147","DOI":"10.1109\/RBME.2009.2034865","article-title":"Histopathological image analysis: a review","volume":"2","author":"Gurcan","year":"2009","journal-title":"IEEE Rev. Biomed. Eng"},{"key":"B13","first-page":"28608","article-title":"\u201cFew-shot object detection with foundation models,\u201d","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Han","year":"2024"},{"key":"B14","article-title":"\u201cDefeating nondeterminism in LLM inference,\u201d","author":"He","year":"2025","journal-title":"Thinking Machines Lab: Connectionism"},{"key":"B15","article-title":"Mask R-CNN","author":"He","year":"2018","journal-title":"arXiv"},{"key":"B16","first-page":"98","article-title":"\u201cFew-shot domain adaptive object detection for microscopic images,\u201d","volume-title":"Proceedings of Medical Image Computing and Computer Assisted Intervention-MICCAI 2024","author":"Inayat","year":"2024"},{"key":"B17","first-page":"4015","article-title":"\u201cSegment anything,\u201d","volume-title":"Proceedings of the IEEE\/CVF International Conference on Computer Vision, pages","author":"Kirillov","year":"2023"},{"key":"B18","doi-asserted-by":"publisher","first-page":"1399","DOI":"10.1007\/s11604-024-01634-z","article-title":"Diagnostic performances of Claude 3 Opus and Claude 3.5 Sonnet from patient history and key images in radiology's 'Diagnosis Please' cases","volume":"42","author":"Kurokawa","year":"2024","journal-title":"Jpn. J. Radiol"},{"key":"B19","first-page":"3041","article-title":"\u201cMask DINO: towards a unified transformer-based framework for object detection and segmentation,\u201d","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Li","year":"2023"},{"key":"B20","doi-asserted-by":"crossref","first-page":"10955","DOI":"10.1109\/CVPR52688.2022.01069","article-title":"\u201cGrounded language-image pre-training,\u201d","volume-title":"2022 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Li","year":"2022"},{"key":"B21","doi-asserted-by":"crossref","first-page":"3491","DOI":"10.1109\/BIBM62325.2024.10821971","article-title":"\u201cCell DINO: end-to-end cell segmentation and tracking with transformer,\u201d","volume-title":"2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","author":"Liao","year":"2024"},{"key":"B22","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1016\/j.media.2017.07.005","article-title":"A survey on deep learning in medical image analysis","volume":"42","author":"Litjens","year":"2017","journal-title":"Med. Image Anal"},{"key":"B23","article-title":"Visual instruction tuning","author":"Liu","year":"2023","journal-title":"arXiv"},{"key":"B24","first-page":"38","article-title":"\u201cGrounding DINO: marrying DINO with grounded pre-training for open-set object detection,\u201d","volume-title":"European Conference on Computer Vision","author":"Liu","year":"2024"},{"key":"B25","doi-asserted-by":"publisher","first-page":"637","DOI":"10.1038\/nmeth.2083","article-title":"Annotated high-throughput microscopy image sets for validation","volume":"9","author":"Ljosa","year":"2012","journal-title":"Nat. Methods"},{"key":"B26","first-page":"1","article-title":"\u201cMicro-bench: a vision-language benchmark for microscopy understanding,\u201d","volume-title":"Proceedings of the 38th International Conference on Neural Information Processing Systems, NIPS '24","author":"Lozano","year":"2024"},{"key":"B27","first-page":"1","article-title":"\u201cRevisiting few-shot object detection with vision-language models,\u201d","volume-title":"Proceedings of the 38th International Conference on Neural Information Processing Systems, NIPS '24","author":"Madan","year":"2025"},{"key":"B28","first-page":"728","article-title":"\u201cSimple open-vocabulary object detection,\u201d","volume-title":"European Conference on Computer Vision","author":"Minderer","year":"2022"},{"key":"B29","article-title":"\u201cDINOv2: Learning robust visual features without supervision,\u201d","author":"Oquab","year":"2024","journal-title":"Transactions on Machine Learning Research. Featured Certification"},{"key":"B30","doi-asserted-by":"crossref","first-page":"2275","DOI":"10.1109\/BIBM49941.2020.9313305","article-title":"\u201cAttention-based transformers for instance segmentation of cells in microstructures,\u201d","volume-title":"2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","author":"Prangemeier","year":"2020"},{"key":"B31","doi-asserted-by":"publisher","first-page":"10","DOI":"10.3390\/bioengineering9100522","article-title":"Deep learning for live cell shape detection and automated AFM navigation","volume":"9","author":"Rade","year":"2022","journal-title":"Bioengineering"},{"key":"B32","article-title":"YOLOv3: an incremental improvement","author":"Redmon","year":"2018","journal-title":"arXiv"},{"key":"B33","first-page":"91","article-title":"\u201cFaster R-CNN: towards real-time object detection with region proposal networks,\u201d","author":"Ren","year":"2015","journal-title":"Proceedings of the 29th International Conference on Neural Information Processing Systems"},{"key":"B34","article-title":"Roboflow100 -VL: a multi-domain object detection benchmark for vision-language models","author":"Robicheaux","year":"2025","journal-title":"arXiv"},{"key":"B35","doi-asserted-by":"publisher","first-page":"221","DOI":"10.1146\/annurev-bioeng-071516-044442","article-title":"Deep learning in medical image analysis","volume":"19","author":"Shen","year":"2017","journal-title":"Annu. Rev. Biomed. Eng"},{"key":"B36","article-title":"BCCD: blood cell count and detection dataset","year":"2017","journal-title":"GitHub Repository"},{"key":"B37","doi-asserted-by":"publisher","first-page":"107157","DOI":"10.1016\/j.vascn.2022.107157","article-title":"High-content analysis and kinetic image cytometry identify toxicity and epigenetic effects of HIV antiretrovirals on human iPSC-neurons and primary neural precursor cells","volume":"114","author":"Smith","year":"2022","journal-title":"J. Pharmacol. Toxicol. Methods"},{"key":"B38","doi-asserted-by":"publisher","first-page":"257","DOI":"10.1007\/s12194-017-0406-5","article-title":"Overview of deep learning in medical imaging","volume":"10","author":"Suzuki","year":"2017","journal-title":"Radiol. Phys. Technol"},{"key":"B39","doi-asserted-by":"crossref","first-page":"1700","DOI":"10.1109\/BigData62323.2024.10825000","article-title":"\u201cBeyond Human Vision: The Role of Large Vision Language Models in Microscope Image Analysis,\u201d","volume-title":"2024 IEEE International Conference on Big Data (BigData)","author":"Verma","year":"2024"},{"key":"B40","doi-asserted-by":"publisher","first-page":"14424","DOI":"10.1073\/pnas.1508524112","article-title":"Advancing biomedical imaging","volume":"112","author":"Weissleder","year":"2015","journal-title":"Proc. Nat. Acad. Sci"},{"key":"B41","first-page":"4452","article-title":"\u201cMulti-modal queried object detection in the wild,\u201d","author":"Xu","year":"2023","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B42","first-page":"106","article-title":"\u201cOpen-vocabulary DETR with conditional matching,\u201d","author":"Zang","year":"2022","journal-title":"Computer Vision"},{"key":"B43","doi-asserted-by":"publisher","first-page":"5625","DOI":"10.1109\/TPAMI.2024.3369699","article-title":"Vision-language models for vision tasks: a survey","volume":"46","author":"Zhang","year":"2024","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell"},{"key":"B44","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2211.09066","article-title":"Teaching algorithmic reasoning via in-context learning","author":"Zhou","year":"2022","journal-title":"arXiv"},{"key":"B45","doi-asserted-by":"publisher","first-page":"820","DOI":"10.1109\/JPROC.2021.3054390","article-title":"A review of deep learning in medical imaging: imaging traits, technology trends, case studies with progress highlights, and future promises","volume":"109","author":"Zhou","year":"2021","journal-title":"Proc. IEEE"},{"key":"B46","doi-asserted-by":"publisher","first-page":"257","DOI":"10.1109\/JPROC.2023.3238524","article-title":"Object detection in 20 years: a survey","volume":"111","author":"Zou","year":"2023","journal-title":"Proc. IEEE"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1761903\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,15]],"date-time":"2026-05-15T05:44:22Z","timestamp":1778823862000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1761903\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,15]]},"references-count":46,"alternative-id":["10.3389\/frai.2026.1761903"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1761903","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,15]]},"article-number":"1761903"}}