MedObvious: Exposing the Medical Moravec's Paradox in VLMs via Clinical Triage

Ufaq Khan; Umair Nawaz; L D M S S Teja; Numaan Saeed; Muhammad Bilal; Yutong Xie; Mohammad Yaqub; Muhammad Haris Khan

cs.CVMar 24, 2026

MedObvious: Exposing the Medical Moravec's Paradox in VLMs via Clinical Triage

Authors: Ufaq Khan, Umair Nawaz, L D M S S Teja, Numaan Saeed, Muhammad Bilal, Yutong Xie, Mohammad Yaqub, Muhammad Haris Khan

Organizations: Mohamed bin Zayed University of Artificial Intelligence, UAE · National Institute of Technology, Silchar · Birmingham City University, UK

Abstract

Vision Language Models (VLMs) are increasingly used for tasks like medical report generation and visual question answering. However, fluent diagnostic text does not guarantee safe visual understanding. In clinical practice, interpretation begins with pre-diagnostic sanity checks: verifying that the input is valid to read (correct modality and anatomy, plausible viewpoint and orientation, and no obvious integrity violations). Existing benchmarks largely assume this step is solved, and therefore miss a critical failure mode: a model can produce plausible narratives even when the input is inconsistent or invalid. We introduce MedObvious, a 1,880-task benchmark that isolates input validation as a set-level consistency capability over small multi-panel image sets: the model must identify whether any panel violates expected coherence. MedObvious spans five progressive tiers, from basic orientation/modality mismatches to clinically motivated anatomy/viewpoint verification and triage-style cues, and includes five evaluation formats to test robustness across interfaces. Evaluating 17 different VLMs, we find that sanity checking remains unreliable: several models hallucinate anomalies on normal (negative-control) inputs, performance degrades when scaling to larger image sets, and measured accuracy varies substantially between multiple-choice and open-ended settings. These results show that pre-diagnostic verification remains unsolved for medical VLMs and should be treated as a distinct, safety-critical capability before deployment.

Figures & tables

Panel of Figure 1 — Figure 1: Qualitative MedObvious examples for pre-diagnostic visual triage. Each column corresponds to a grid in (a)–(c). We report the task question, ground truth, and predictions from representative VLMs.

Figure 2: MedObvious overview. ( A ) Five progressive tiers (T1–T5). ( B ) Construction: multi-view studies are abstracted into small grids with either a single outlier or a negative-control (no outlier). ( C ) Five evaluation protocols: Detection (MCQ/Open), Referring (MCQ/Open), and Visual Referring (Yes/No).

Tier	Name	Grid	#Mod.	Tasks	Pos.	Neg.	Clinical analog
T1	Foundation	$2\times 2$	2	440	275	165	Basic QC (orientation, modality)
T2	Diversity	$2\times 2$	2	480	300	180	Robustness across sources
T3	Scaling	$3\times 3$	4	360	225	135	Multi-view/-slice comparison
T4	Semantics	$2\times 2$	3	320	200	120	Anatomy/viewpoint verification
T5	Triage	Mixed	4	280	175	105	“Stop and verify” safety cues
Total			4	1,880	1,175	705

Table 1: MedObvious composition. Tiers increase in clinical specificity.

Model	Task-A	Task-B	Task-C	Task-D	Task-E	Pos( $+$ )	Neg( $-$ )	Avg
General open-source VLMs
LLaVA-1.5-7B	40.4	22.3	22.1	35.7	50.0	37.5	31.9	34.3
Qwen2-VL-7B	32.3	49.5	72.7	25.1	53.1	56.3	28.7	45.4
Qwen2.5-VL-7B	58.7	82.1	75.3	29.7	65.1	60.9	70.7	63.2
Qwen3-VL-8B	31.2	32.7	80.8	38.7	52.9	68.9	2.9	44.0
InternVL2.5-8B	56.3	56.3	69.7	26.8	51.7	59.7	42.5	51.9

Table 2: MedObvious: Per-task accuracy (%). Task A–E correspond to Detection MCQ, Detection Open, Referring MCQ, Referring Open, and Visual Referring, respectively. “Positive (Pos)”/“Negative (Neg)” indicate accuracy on anomalous/non-anomalous samples. Best results per task are bolded .

VLM Evaluation

Vision-language model evaluation measures multimodal systems’ perception, reasoning, grounding, robustness, and reliability across controlled and real-world tasks.

See all 957 papers

Consistent but Dangerous: Per-Sample Safety Classification Reveals False Reliability in Medical Vision-Language Models
Mar 22, 2026Binesh Sadanandan, Vahid BehzadanVLM Evaluation Medical VLMs
SAIL Lab, University of New Haven, West Haven, CT, USA
MedVIGIL: Evaluating Trustworthy Medical VLMs Under Broken Visual Evidence
May 8, 2026Hanqi Jiang, Junhao Chen, Mingyu Kang +12VLM Evaluation VLM Robustness
University of Georgia · Chungbuk National University · Chungnam National University Hospital +5
Is There Knowledge Left to Extract? Evidence of Fragility in Medically Fine-Tuned Vision-Language Models
Apr 10, 2026Oliver McLaughlin, Daniel Shubin, Carsten Eickhoff +3Medical VLMs VLM Reasoning
Brown University · University of Washington · University of Tübingen +1

Mar 22, 2026J/K move · Enter open · S save

Consistent but Dangerous: Per-Sample Safety Classification Reveals False Reliability in Medical Vision-Language Models

Binesh Sadanandan, Vahid Behzadan

SAIL Lab, University of New Haven, West Haven, CT, USA

VLM Evaluation Medical VLMs Faithfulness in VLMs Prompt Sensitivity

Open paper AskPDF arXiv

Consistency under paraphrase, the property that semantically equivalent prompts yield identical predictions, is increasingly used as a proxy for reliability when deploying medical vision-language models (VLMs). We show this proxy is fundamentally flawed: a model can achieve perfect consistency by relying on text patterns rather than the input image. We introduce a four-quadrant per-sample safety taxonomy that jointly evaluates consistency (stable predictions across paraphrased prompts) and image reliance (predictions that change when the image is removed). Samples are classified as Ideal (consistent and image-reliant), Fragile (inconsistent but image-reliant), Dangerous (consistent but not image-reliant), or Worst (inconsistent and not image-reliant). Evaluating five medical VLM configurations across two chest X-ray datasets (MIMIC-CXR, PadChest), we find that LoRA fine-tuning dramatically reduces flip rates but shifts a majority of samples into the Dangerous quadrant: LLaVA-Rad Base achieves a 1.5% flip rate on PadChest while 98.5% of its samples are Dangerous. Critically, Dangerous samples exhibit high accuracy (up to 99.6%) and low entropy, making them invisible to standard confidence-based screening. We observe a negative correlation between flip rate and Dangerous fraction (r = -0.89, n=10) and recommend that deployment evaluations always pair consistency checks with a text-only baseline: a single additional forward pass that exposes the false reliability trap.

Abstract

VLM Evaluation

Explore similar work

Consistent but Dangerous: Per-Sample Safety Classification Reveals False Reliability in Medical Vision-Language Models

MedVIGIL: Evaluating Trustworthy Medical VLMs Under Broken Visual Evidence

Is There Knowledge Left to Extract? Evidence of Fragility in Medically Fine-Tuned Vision-Language Models