Anatomy-aware Fine-grained Multimodal Fusion for Laryngopharyngeal Cancer T-Staging Prediction Using CT and Radiology Report

Xingyue Zhao; Yanzhou Su; Fang Zhang; Zhanghexuan Ji; Yirui Wang; Dazhou Guo; Sibo Ju; Yuehua Cheng; Yuzhen Chen; Ming Feng; Le Lu; Tsung-Ying Ho; Jian Wang; Dakai Jin; Na Shen

arXiv · 2610.06837

Anatomy-aware Fine-grained Multimodal Fusion for Laryngopharyngeal Cancer T-Staging Prediction Using CT and Radiology Report

Xingyue Zhao·Yanzhou Su·Fang Zhang·Zhanghexuan Ji·Yirui Wang·Dazhou Guo·Sibo Ju·Yuehua Cheng·Yuzhen Chen·Ming Feng·Le Lu·Tsung-Ying Ho·Jian Wang·Dakai Jin·Na Shen

View original source ↗Read source document ↗Export metadata ↓

Abstract

Accurate T-staging is crucial for guiding personalized treatment strategies for laryngopharyngeal cancer. However, current clinical practice relies on invasive biopsy procedures, whereas CT-based staging remains challenging due to the complex patterns of tumor invasion. Recent computer-aided approaches face two key challenges: 1) Structural relationship modeling: existing methods underrepresent anatomically structured patterns of tumor invasion, as they either process whole CT volumes without tumor-specific anatomical constraints or rely on labor-intensive tumor segmentation. 2) Fine-grained cross-modal alignment: while radiology reports contain organ-specific invasion details, current methods that apply global feature fusion struggle to accurately align individual anatomical structures with their corresponding textual descriptions. To address these issues, we propose an anatomy-aware multimodal framework that integrates organ-level CT context and radiology reports into a unified representation for laryngopharyngeal T-staging. The framework first constructs an Anatomy-Structured Organ Graph (AOG) that captures invasion patterns between primary sites and surrounding organs, then performs Organ-Anchored Cross-Modal Alignment (OCA) so that each organ node aggregates textual evidence from the radiology report, and finally refines this graph representation by injecting organ-specific invasion cues extracted from the report via Report-Enhanced Graph-Refinement (REG), yielding a multimodal organ graph that combines spatial and textual evidence. Extensive experiments demonstrate that the proposed framework achieves superior performance in T-staging of laryngopharyngeal cancer.

Explore related subjects

cs.CV

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Xingyue Zhao, Yanzhou Su, Fang Zhang, Zhanghexuan Ji, Yirui Wang, Dazhou Guo, Sibo Ju, Yuehua Cheng, Yuzhen Chen, Ming Feng, Le Lu, Tsung-Ying Ho, Jian Wang, Dakai Jin, Na Shen. 2026-10-05. Anatomy-aware Fine-grained Multimodal Fusion for Laryngopharyngeal Cancer T-Staging Prediction Using CT and Radiology Report. https://arxiv.org/abs/2610.06837

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Environmental Change Detection for Real-World Change Analysis

Scene Change Detection (SCD) evaluates changes using predefined query-reference (i.e., present-past) image pairs. However, this formulation overlooks a critical dependency: the corresponding query-reference pair is assumed to be prepared in advance. In real-world applications, such as mobile robots, future query views cannot be known in advance, and thus their corresponding reference images cannot be predefined. To remove this dependency and push change detection toward more practical applications, we introduce Environmental Change Detection (ECD). A key aspect of ECD is to avoid unrealistically predefined and aligned query-reference pairs and instead retrieve environmental cues from an uncurated image database of reference scenes. To tackle this new challenging task, we additionally introduce an initial solution that enables change detection under unknown and imperfect query-reference conditions. The main idea of our solution is to retrieve multiple reference candidates and aggregate semantically rich representations for change detection. We further construct ECD benchmark sets by reformulating three standard change detection datasets. Extensive experimental results demonstrate the efficacy of our solution in both ECD and SCD.

cs.CV↗

arXiv2026

VLANeXt Family: A Systematic Study of VLA Models from Core Recipes to Emerging Paradigms

Following the rise of large foundation models, Vision-Language-Action models (VLAs) emerged, leveraging strong visual and language understanding from Vision-Language Models for general-purpose policy learning. Yet, the current VLA landscape remains fragmented and exploratory. Although many groups have proposed their own VLA models, inconsistencies in training protocols and evaluation settings make it difficult to identify which design choices truly matter. To bring structure to this evolving space, we reexamine the VLA design space under a unified framework and evaluation setup. Starting from a simple VLA baseline similar to RT-2, which is the origin of VLA, we systematically dissect design choices along three dimensions: foundational components, perception essentials, and action modeling perspectives. From this study, we distill 12 key findings that together form a practical recipe for building strong VLA models. The outcome of this exploration is a simple yet effective model, VLANeXt. It outperforms the state-of-the-art methods on the LIBERO and LIBERO-plus benchmarks and demonstrates strong performance in real-world experiments. Beyond identifying the core recipe, we further ask how far these design principles extend to the emerging paradigms in VLAs. We thus expand VLANeXt along several emerging directions, including model scaling, latent-action pretraining, latent predictive representation learning, and world action modeling. These studies give rise to the VLANeXt family, spanning compact and scaled VLA variants, latent-action models, JEPA-style predictive models, and World Action Models. Our results show that the core recipe provides a strong foundation across different model scales and emerging paradigms.

cs.CV↗

arXiv2026

Attention at Rest Stays at Rest: Breaking Visual Inertia to Mitigate Relation Hallucinations

While multimodal large language models demonstrate strong entity-level perception, faithfully grounding relational interactions between objects remains a persistent challenge. Although conventional visual grounding techniques attempt to resolve hallucinations by amplifying visual attention, strengthening overall visual signals fails to reliably correct relational errors. Tracing visual attention in relation descriptions reveals that correct responses tend to dynamically shift focus across regions, whereas hallucinated responses often linger on previously dominant evidence, exhibiting an undesirable \textit{visual inertia}. Further analysis shows that relation-prediction performance steadily deteriorates as more previous-step visual attention is carried into the current decoding step. We therefore introduce Inertia-aware Visual Excitation (IVE), an MLLM decoding method that dynamically recalibrates visual values using token-level attention history. By contrasting current attention against recent moving averages, IVE separates emergent tokens with rising relevance from persistently dominant inertia tokens, selectively reinforcing newly needed evidence while mildly attenuating contributions from repeatedly attended regions. Across three MLLMs and decoding strategies, IVE reduces relation hallucinations while preserving broader multimodal performance.

cs.CV↗