MedJev: A System One Vision-Language Decision Model for Fast and Effective Medical Image Understanding
Summary
The paper introduces MedJev, a System One vision-language decision model that maps medical images, questions, and candidate answers directly to decisions in a single forward pass without autoregressive generation. Built on Qwen3.5-4B, MedJev-4B achieves a 7.93% absolute accuracy gain and 6.04x end-to-end speedup, reaching state-of-the-art performance among 4B–9B general and medical VLMs.
View Cached Full Text
Cached at: 11/19/26, 03:37 AM
Paper page - MedJev: A System One Vision-Language Decision Model for Fast and Effective Medical Image Understanding
Source: https://huggingface.co/papers/2611.20992
Abstract
Medicalimageunderstandingwithvision-languagemodels(VLMs)typicallyreliesonautoregressivelygeneratingreasoningandanswertokens.Theseunnecessarytokensmaketheinferencecostscalewiththeresponselengthandforcetheanswertobeparsedfromfree-formtext,incurringavoidablelatencyanddecisionerrors.Tothisend,weintroduceMedJev,asystemonevision-languagedecisionmodelthatmapsmedicalimages,questions,andcandidateanswersdirectlytodecisionsinasinglemultimodalforwardpass,withoutanyautoregressivegenerationprocess.Forfastinference,MedJevremovestheupperlanguagelayersandthegenerativeoutputpathway,whilekeepingthefullvisionpathwayandallvisualtokens.Onthisshortenedbackbone,adecisionheadscoreseachcandidatebypretrainedquery--keymatchingandrefinesthescoreswithalightweightmultilayerperceptron(MLP)residual,couplingthepretrainedmatchingstructurewithtask-specificnonlineardiscrimination.BuiltonQwen3.5-4B,MedJev-4Bimprovesbothaccuracyandspeedovertheoriginalmodeloneachoffourmedicalimageclassificationandvisualquestionansweringbenchmarks,withanabsoluteaccuracygainof7.93\%anda6.04timesend-to-endspeedup.Furthercomparisonwith20cutting-edgemodelsshowsthatMedJev-4Bachievesstate-of-the-artperformanceandinferencespeedamong4B--9BgeneralandmedicalVLMs,indicatingthatdirectmultimodaldecisionmakingisaneffectiveroutetofastandeffectivemedicalimageunderstanding.
View arXiv pageView PDFGitHub42Add to collection
Get this paper in your agent:
hf papers read 2611\.20992
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2611.20992 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2611.20992 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2611.20992 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
OmniMed-Jev: Calibrating LVLM Confidence for Trustworthy Medical Multimodal Decisions via System One
OmniMed-Jev reframes medical LVLM outputs as distributions over a runtime-supplied candidate set, cutting calibration error by up to an order of magnitude and reliability error by two versus a generative baseline while keeping point-prediction performance comparable (except counting).
MedJev: Evaluating Zero-Shot System-One Decision Making in Medicine
MedJev 是一个可在笔记本本地运行的开源 System-One 医疗模型,在十项医学基准上进行零样本评测,性能接近 GPT-4,并提出两参数再校准方法显著改善患者级概率校准。
A System 1 Decision Model for Medical Multiple-Choice Questions Built on Frozen Language Models
This paper proposes a System 1 decision model for medical multiple-choice questions that trains only small projection heads and a temperature on top of frozen language models, achieving up to 46.0% accuracy on MedMCQA as a cheap, fast baseline.
🚀 Jev-style decisions for IMAGES are here: Fovea-4B, an open 4B model leading on ImageJev tasks (Apache-2.0)
Fovea-4B is an open-weight 4B vision model that brings Jev-style typed decision contracts to images, leading ImageJev-style tasks with 80.6 chance-corrected accuracy and strong calibration (ECE 0.035), released under Apache-2.0.
SearchJev: A Fast and Calibrated System-1 Model for Search Agents
SearchJev introduces a fast, calibrated System-1 model for search agents that scores legal decisions without autoregressive generation, using Soft-Label Learning for Calibrated Decisions (SLCD) to reduce latency and improve confidence reliability. A dual-system agent combining SearchJev with a System-2 model achieves up to 4.7x speedups on BrowseComp-Plus while raising answer accuracy from 45% to 54%.