Institution profile

University of Missouri-Kansas City

Academic institutionnorthamerica · us
Official website
Research library26linked papers
Opportunities0open roles
Selected work

Representative Papers

Predicting When to Trust Vision-Language Models for Spatial Reasoning

Jan 14, 2026

This work addresses the systematic errors and unreliable confidence estimation of vision-language models (VLMs) in spatial reasoning tasks, which hinder their deployment in safety-critical applications. The authors propose a novel confidence estimation framework grounded in external geometric verification, eschewing conventional text-based self-evaluation. Their approach integrates four visual signals—object detection, geometric alignment, spatial ambiguity, and internal VLM uncertainty—and combines them via a gradient-boosting model to predict reliability. Evaluated on BLIP-2 and CLIP, the method achieves AUROC scores of 0.674 and 0.583, respectively, representing improvements of 34.0% and 16.1% over baselines. At 60% target accuracy, it attains a coverage rate of 61.9%—more than twice that of the baseline—and boosts scene graph construction accuracy from 52.1% to 78.3%, demonstrating strong cross-architecture generalization.

2 citationsRead paper

MATEX: Multi-scale Attention and Text-guided Explainability of Medical Vision-Language Models

Jan 16, 2026

This work addresses the limited interpretability of existing medical vision-language models, which stems from their lack of anatomical grounding, imprecise spatial localization, and coarse attention granularity. To overcome these limitations, the authors propose a novel approach that integrates multi-scale attention rollout, text-guided spatial priors, and inter-layer consistency analysis to generate anatomically aligned, high-precision, and stable gradient attribution maps. Experimental results on the MS-CXR dataset demonstrate that the proposed method outperforms the current state-of-the-art M2IB model in both spatial localization accuracy and alignment with expert-annotated lesion regions, thereby significantly enhancing the clinical interpretability of medical AI systems.

1 citationsRead paper

Clinical Knowledge Graph Construction and Evaluation with Multi-LLMs via Retrieval-Augmented Generation

Jan 05, 2026arXiv.org

This work addresses the limitations of existing approaches in constructing oncology knowledge graphs from unstructured clinical text, which often lack effective fact verification and semantic consistency. The authors propose an end-to-end KG-RAG framework that integrates multi-agent prompt engineering, retrieval-augmented generation, and ontology-aligned RDF/OWL semantic modeling to directly extract entities, attributes, and relations. To mitigate hallucination and enhance semantic fidelity, the method incorporates an entropy-based uncertainty scoring mechanism and a multi-LLM consensus strategy. Notably, it enables gold-standard-free, self-supervised continuous refinement. Evaluated on PDAC and BRCA patient cohorts, the resulting knowledge graphs demonstrate high clinical credibility, SPARQL compatibility, and significant improvements over baseline methods in precision, relevance, and ontological compliance.

1 citationsRead paper
Recent publications

Latest Papers

Adapting Vision Foundation Models with Cascaded Semantics

Aug 05, 2026

Existing visual prompt tuning methods rely on random initialization and overlook semantic priors, thereby limiting the adaptability of pretrained Vision Transformers (ViTs) to downstream tasks. This work proposes a cascaded prompt tuning framework that, for the first time, integrates handcrafted input-space priors—such as color, texture, and shape—with instance-aware self-attention semantics in feature space. By synergistically combining these complementary dual-level priors, the framework effectively guides prompt learning. Remarkably, the method achieves state-of-the-art performance across 34 image classification benchmarks while tuning only 0.74% of the ViT parameters, demonstrating both high efficiency and strong adaptation capability for downstream tasks.

0 citationsRead paper

SEMIR: Topology-Preserving Graph Minors for Thin-Structure Segmentation

Jun 22, 2026

This work addresses the fragmentation problem in segmenting fine structures—such as wires, cracks, and lane markings—caused by traditional pixel-wise representations that disrupt topological connectivity. To preserve structural continuity, the authors propose a topology-preserving, parameterized graph minor representation that compresses the input image into super-nodes via a boundary-aligned contraction criterion. This approach achieves substantial dimensionality reduction while rigorously maintaining the connectivity of fine structures. A lightweight graph neural network is then employed for classification, followed by bidirectional pixel-to-graph mapping to enable full-resolution inference. Evaluated on TTPLA, CrackSeg9k, and SkyScapes Lane datasets, the method matches or surpasses state-of-the-art domain-specific approaches in Dice, IoU, and Boundary F1 scores, while reducing mask fragmentation by at least 4.6×.

0 citationsRead paper