arXiv subjects

Tianze Yu

Publications and source records attributed to Tianze Yu.

10 recordsLinked to original sources

arXiv2026

How Do Medical MLLMs Fail? A Study on Visual Grounding in Medical Images

Generalist multimodal large language models (MLLMs) have achieved impressive performance across a wide range of vision-language tasks. However, their performance on medical tasks, particularly in zero-shot settings where generalization is critical, remains suboptimal. A key research gap is the limited understanding of why medical MLLMs underperform in medical image interpretation. In this work, we present a pioneering systematic investigation into the visual grounding capabilities of state-of-the-art medical MLLMs. To disentangle visual grounding from semantic grounding, we design VGMED, a novel evaluation dataset developed with expert clinical guidance, explicitly assessing the visual grounding capability of medical MLLMs. We introduce new quantitative metrics and conduct detailed qualitative analyses. Our study across eight state-of-the-art (SOTA) medical MLLMs validates that they often fail to ground their predictions in clinically relevant image regions. We note that this finding is specific to medical image analysis; in contrast, prior work has shown that MLLMs are capable of grounding their predictions in the correct image regions when applied to natural scene images. Motivated by these findings, we propose VGRefine, a simple yet effective inference-time method that refines attention distribution to improve visual grounding in medical settings. Our approach achieves SOTA performance across 6 diverse Med-VQA benchmarks (over 110K VQA samples from 8 imaging modalities) without requiring additional training or external expert models. Overall, our work, for the first time, systematically validates inadequate visual grounding as one of the key contributing factors for medical MLLMs' under-performance. Additional experiments are included in the Supp.

cs.CV↗

arXiv2025

MUSON: A Reasoning-oriented Multimodal Dataset for Socially Compliant Navigation in Urban Environments

Socially compliant navigation requires structured reasoning about dynamic pedestrians and physical constraints to ensure safe and interpretable decisions. Vision-language models (VLMs) provide a promising foundation for this task because they can integrate visual observations with language-based social knowledge. However, existing untuned VLMs still struggle to reliably understand fine-grained social norms, making task-specific fine-tuning essential. At the same time, no large-scale egocentric dataset is available this task. To address these challenges, we introduce MUSON, a multimodal dataset for short-horizon social navigation containing 10,110 egocentric samples collected across diverse indoor and outdoor social scenes. MUSON adopts a structured five-step chain-of-thought annotation framework comprising perception, prediction, reasoning, action, and explanation. It explicitly models static physical constraints and employs a standardized six-action decision space. Compared with existing social-navigation datasets, MUSON provides consistent annotations for reasoning, actions, and explanations. We evaluate ten representative small-to-medium VLMs on MUSON. Qwen3-VL-8B achieves the strongest decision-level performance, attaining the highest action accuracy of 0.7765 and Macro-F1 score of 0.7490, as well as the lowest collision rate of 0.0609. These results demonstrate that MUSON is an effective and reusable benchmark for advancing socially compliant navigation. The dataset is publicly available at https://github.com/MUSON-dataset/MUSON/releases/tag/v1.0.

cs.CV↗

arXiv2025

ASLSL: Adaptive shared latent structure learning with incomplete multi-modal physiological data for multi-dimensional emotional feature selection

Recently, multi-modal physiological signals based emotion recognition has garnered increasing attention in the field of brain-computer interfaces. Nevertheness, the associated multi-modal physiological features are often high-dimensional and inevitably include irrelevant, redundant, and noisy representation, which can easily lead to overfitting, poor performance, and high computational complexity in emotion classifiers. Feature selection has been widely applied to address these challenges. However, previous studies generally assumed that multi-modal physiological data are complete, whereas in reality, the data are often incomplete due to the openness of the acquisition and operational environment. For example, a part of samples are available in several modalities but not in others. To address this issue, we propose a novel method for incomplete multi-modal physiological signal feature selection called adaptive shared latent structure learning (ASLSL). Based on the property that similar features share similar emotional labels, ASLSL employs adaptive shared latent structure learning to explore a common latent space shared for incomplete multi-modal physiological signals and multi-dimensional emotional labels, thereby mitigating the impact of missing information and mining consensus information. Two most popular multi-modal physiological emotion datasets (DEAP and DREAMER) with multi-dimensional emotional labels were utilized to compare the performance between compare ASLSL and seventeen feature selection methods. Comprehensive experimental results on these datasets demonstrate the effectiveness of ASLSL.

cs.HC↗

arXiv2025

ADSEL: Adaptive dual self-expression learning for EEG feature selection via incomplete multi-dimensional emotional tagging

EEG based multi-dimension emotion recognition has attracted substantial research interest in human computer interfaces. However, the high dimensionality of EEG features, coupled with limited sample sizes, frequently leads to classifier overfitting and high computational complexity. Feature selection constitutes a critical strategy for mitigating these challenges. Most existing EEG feature selection methods assume complete multi-dimensional emotion labels. In practice, open acquisition environment, and the inherent subjectivity of emotion perception often result in incomplete label data, which can compromise model generalization. Additionally, existing feature selection methods for handling incomplete multi-dimensional labels primarily focus on correlations among various dimensions during label recovery, neglecting the correlation between samples in the label space and their interaction with various dimensions. To address these issues, we propose a novel incomplete multi-dimensional feature selection algorithm for EEG-based emotion recognition. The proposed method integrates an adaptive dual self-expression learning (ADSEL) with least squares regression. ADSEL establishes a bidirectional pathway between sample-level and dimension-level self-expression learning processes within the label space. It could facilitate the cross-sharing of learned information between these processes, enabling the simultaneous exploitation of effective information across both samples and dimensions for label reconstruction. Consequently, ADSEL could enhances label recovery accuracy and effectively identifies the optimal EEG feature subset for multi-dimensional emotion recognition.

cs.HC↗

arXiv2025

FDC-Net: Rethinking the association between EEG artifact removal and multi-dimensional affective computing

Electroencephalogram (EEG)-based emotion recognition holds significant value in affective computing and brain-computer interfaces. However, in practical applications, EEG recordings are susceptible to the effects of various physiological artifacts. Current approaches typically treat denoising and emotion recognition as independent tasks using cascaded architectures, which not only leads to error accumulation, but also fails to exploit potential synergies between these tasks. Moreover, conventional EEG-based emotion recognition models often rely on the idealized assumption of "perfectly denoised data", lacking a systematic design for noise robustness. To address these challenges, a novel framework that deeply couples denoising and emotion recognition tasks is proposed for end-to-end noise-robust emotion recognition, termed as Feedback-Driven Collaborative Network for Denoising-Classification Nexus (FDC-Net). Our primary innovation lies in establishing a dynamic collaborative mechanism between artifact removal and emotion recognition through: (1) bidirectional gradient propagation with joint optimization strategies; (2) a gated attention mechanism integrated with frequency-adaptive Transformer using learnable band-position encoding. Two most popular EEG-based emotion datasets (DEAP and DREAMER) with multi-dimensional emotional labels were employed to compare the artifact removal and emotion recognition performance between FDC-Net and nine state-of-the-art methods. In terms of the denoising task, FDC-Net obtains a maximum correlation coefficient (CC) value of 96.30% on DEAP and a maximum CC value of 90.31% on DREAMER. In terms of the emotion recognition task under physiological artifact interference, FDC-Net achieves emotion recognition accuracies of 82.3+7.1% on DEAP and 88.1+0.8% on DREAMER.

cs.HC↗

arXiv2025

Guiding polaritonic energy and momentum through two-dimensional Bravais lattices

The strong exciton absorption in monolayer transition metal dichalcogenides provides a promising platform for studying polaritons with tunable dispersions, which are crucial for controlling polaritonic energy and momentum, but remain underexplored. In this work, monolayer MoS$_2$ is coupled with a Fabry-P\'erot microcavity to form polaritons. Five types of Bravais lattices with sub-wavelength periods, based on polymethyl methacrylate (PMMA) nanopillars, are intentionally designed. The energy overlap between the periodic PMMA scattering wave and the polariton establishes a coupling channel that controls the directional flow of polaritonic energy, as demonstrated through angle-resolved reflectance measurements. Back-space image measurements further demonstrate that the dispersion in reciprocal space can be directly and manually tuned, allowing for control over their number and their positions. The coupling between the polariton and PMMA scattering wave is further demonstrated by analyzing the reflectance using the two-port two-mode model. The symmetries of 2D Bravais lattices allow the angle between energy and momentum flow to vary widely, from 90{\deg}, 60{\deg}, 45{\deg}, and 30{\deg} to arbitrary values. By adjusting the lattice vector lengths, the position of the dispersion branch in a specific direction can be fine-tuned, enabling full-range control over polariton dispersion. This work presents the first theoretical and experimental demonstrations of guiding the direction of polaritonic energy and momentum through Bravais lattice design.

physics.optics↗

arXiv2024

Integrating Clinical Knowledge Graphs and Gradient-Based Neural Systems for Enhanced Melanoma Diagnosis via the 7-Point Checklist

The 7-point checklist (7PCL) is a widely used diagnostic tool in dermoscopy for identifying malignant melanoma by assigning point values to seven specific attributes. However, the traditional 7PCL is limited to distinguishing between malignant melanoma and melanocytic Nevi, and falls short in scenarios where multiple skin diseases with appearances similar to melanoma coexist. To address this limitation, we propose a novel diagnostic framework that integrates a clinical knowledge-based topological graph (CKTG) with a gradient diagnostic strategy featuring a data-driven weighting system (GD-DDW). The CKTG captures both the internal and external relationships among the 7PCL attributes, while the GD-DDW emulates dermatologists' diagnostic processes, prioritizing visual observation before making predictions. Additionally, we introduce a multimodal feature extraction approach leveraging a dual-attention mechanism to enhance feature extraction through cross-modal interaction and unimodal collaboration. This method incorporates meta-information to uncover interactions between clinical data and image features, ensuring more accurate and robust predictions. Our approach, evaluated on the EDRA dataset, achieved an average AUC of 88.6%, demonstrating superior performance in melanoma detection and feature prediction. This integrated system provides data-driven benchmarks for clinicians, significantly enhancing the precision of melanoma diagnosis.

cs.CV↗

arXiv2024

Extremely long transverse optical needle focus for reflective metalens enabled by monolayer MoS$_2$

Line-scan mode facilitates fast-speed and high-throughput imaging with developing a suitable optical transverse needle focus. Metasurface with periodic structures such as diffractive rings, ellipses, and gratings could enable discrete focus evolving into line focus under momentum conservation, but still face the challenge of extremely low light power utilization brought by inevitably multiple high-order diffractions. In addition, the designed focus requires the selection of particular optical functional materials. High dielectric constants in atomic transition metal dichalcogenides make significant phase modulation by bringing phase singularity at zero-reflection possible. However, no light power is available for use at zero-reflection and a balance between phase and amplitude modulation is needed. In this work, above issues are simultaneously solved by designing a monolayer MoS2 based Fresnel strip structure. An optical needle primary focus with a transverse length of 40 {\mu}m (~80 {\lambda}) is obtained, which is the longest value recorded so far, together with a sub-diffraction-limited lateral spot and a broad working wavelength range. This specially developed structure not only concentrates light power in primary diffraction by breaking restriction of momentum conservation, but also guarantees a consistent phase across different strips. The novel optical manipulation way provided here together with the longer focus length for flat optics will show promising applications in biology, oncology, nanofabrication, energy harvesting, and optical information processing.

physics.optics↗

arXiv2021

Rethinking Crowdsourcing Annotation: Partial Annotation with Salient Labels for Multi-Label Image Classification

Annotated images are required for both supervised model training and evaluation in image classification. Manually annotating images is arduous and expensive, especially for multi-labeled images. A recent trend for conducting such laboursome annotation tasks is through crowdsourcing, where images are annotated by volunteers or paid workers online (e.g., workers of Amazon Mechanical Turk) from scratch. However, the quality of crowdsourcing image annotations cannot be guaranteed, and incompleteness and incorrectness are two major concerns for crowdsourcing annotations. To address such concerns, we have a rethinking of crowdsourcing annotations: Our simple hypothesis is that if the annotators only partially annotate multi-label images with salient labels they are confident in, there will be fewer annotation errors and annotators will spend less time on uncertain labels. As a pleasant surprise, with the same annotation budget, we show a multi-label image classifier supervised by images with salient annotations can outperform models supervised by fully annotated images. Our method contributions are 2-fold: An active learning way is proposed to acquire salient labels for multi-label images; and a novel Adaptive Temperature Associated Model (ATAM) specifically using partial annotations is proposed for multi-label image classification. We conduct experiments on practical crowdsourcing data, the Open Street Map (OSM) dataset and benchmark dataset COCO 2014. When compared with state-of-the-art classification methods trained on fully annotated images, the proposed ATAM can achieve higher accuracy. The proposed idea is promising for crowdsourcing data annotation. Our code will be publicly available.

cs.CV↗

arXiv2021

SCIDA: Self-Correction Integrated Domain Adaptation from Single- to Multi-label Aerial Images

Most publicly available datasets for image classification are with single labels, while images are inherently multi-labeled in our daily life. Such an annotation gap makes many pre-trained single-label classification models fail in practical scenarios. This annotation issue is more concerned for aerial images: Aerial data collected from sensors naturally cover a relatively large land area with multiple labels, while annotated aerial datasets, which are publicly available (e.g., UCM, AID), are single-labeled. As manually annotating multi-label aerial images would be time/labor-consuming, we propose a novel self-correction integrated domain adaptation (SCIDA) method for automatic multi-label learning. SCIDA is weakly supervised, i.e., automatically learning the multi-label image classification model from using massive, publicly available single-label images. To achieve this goal, we propose a novel Label-Wise self-Correction (LWC) module to better explore underlying label correlations. This module also makes the unsupervised domain adaptation (UDA) from single- to multi-label data possible. For model training, the proposed model only uses single-label information yet requires no prior knowledge of multi-labeled data; and it predicts labels for multi-label aerial images. In our experiments, trained with single-labeled MAI-AID-s and MAI-UCM-s datasets, the proposed model is tested directly on our collected Multi-scene Aerial Image (MAI) dataset.

cs.CV↗