Xinyu Ma

Noise Improves Multimodal Machine Translation: Rethinking the Role of Visual Context

Xinyu Ma, Jun Rao, Xuebo Liu

Multimodal Machine Translation (MMT) has long been assumed to outperform traditional text-only MT by leveraging visual information. However, recent studies challenge this assumption, showing that MMT models perform similarly even when tested without images or with mismatched images. This raises fundamental questions …

cn (code pays fourni par la source)

Global Fused Multiresolution Semantic Interactive Network for Low-Contrast Infrared Iceberg Detection

Xinyu Ma, Yang Li, Fanglin Chen, Fuhai Guo et autres

Ice floes and icebergs pose a deadly threat to ships. Ensuring safe navigation by timely detecting these hazards through shipborne detection systems represents a potent solution. Infrared iceberg detection stands out due to its exceptional visual capability and immunity to light and …

cn (code pays fourni par la source)

Cross-Attention Adaptive Feature Pyramid Network with Uncertainty Boundary Modeling for Mass Detection in Digital Breast Tomosynthesis

Xinyu Ma, Haotian Sun, Gang Yuan, Yufei Tang et autres

Computer-aided detection (CADe) of masses in digital breast tomosynthesis (DBT) is crucial for early breast cancer diagnosis. However, the variability in the size and morphology of breast masses and their resemblance to surrounding tissues present significant challenges. Current CNN-based CADe methods, particularly …

cn (code pays fourni par la source)

Adaptive Multiscale Enhanced Interaction Network for Infrared Iceberg Detection

Xinyu Ma, Yang Li, Fanglin Chen, Fuhai Guo et autres

Icebergs threaten the navigation safety of ships in the Arctic, small icebergs are difficult to detect in harsh sea conditions. Timely detection of these targets is crucial for aiding ships in navigational decisions. Infrared detection technology is an effective solution for detecting …

cn (code pays fourni par la source)

Sparse Cascade: Multi-stage cascade fusion and shape dictionary guided network for segmentation of microscopic cytology images

Ao Liu, Yuan Wang, Jie Ding, Xinyu Ma et autres

Microscopic cytology image instance segmentation is critical for clinical diagnosis, treatment planning, and biomedical research. However, challenges such as cell overlapping, external interferences, and irregular shapes hinder accurate segmentation. In this paper, a microscopic cytology image instance segmentation algorithm based on a …

cn (code pays fourni par la source)

LoRA Dropout as a Sparsity Regularizer for Overfitting Control

Xinyu Ma, Xu Chu, Yujie Jin, Zhibang Yang et autres

Parameter-efficient fine-tuning methods, represented by LoRA, play an essential role in adapting large-scale pre-trained models to downstream tasks. However, fine-tuning LoRA-series models also faces the risk of overfitting on the training dataset, and yet there's still a lack of theoretical guidance and …

DB-DCAFN: dual-branch deformable cross-attention fusion network for bacterial segmentation

Jingkun Wang, Xinyu Ma, Long Cao, Yilin Leng et autres

Sputum smear tests are critical for the diagnosis of respiratory diseases. Automatic segmentation of bacteria from sputum smear images is important for improving diagnostic efficiency. However, this remains a challenging task owing to the high interclass similarity among different categories of bacteria …

cn (code pays fourni par la source)

Edge-Aided Multiscale Context Network for Infrared Small Target Detection

Xinyu Ma, Yang Li

Infrared small target detection technology has been widely used in various fields. Infrared small target usually shows the characteristics of low gray values and a lack of texture information in complex scenes, which makes it difficult to accurately segment the boundary of …

cn (code pays fourni par la source)

Fused Gromov-Wasserstein Graph Mixup for Graph-level Classifications

Xinyu Ma, Xu Chu, Yasha Wang, Yang Lin et autres