Searcharxiv⌕ Search

arXiv subjects

Dehua Chen

Publications and source records attributed to Dehua Chen.

3 recordsLinked to original sources

DroneSegEdit: Adapting Pretrained Generative Editors for UAV Segmentation with Language-Grounded Verification

Recent work has shown that pretrained image generators can support diverse visual tasks. Yet directly applying them to language-guided unmanned aerial vehicle (UAV) segmentation exposes a key limitation: a plausible mask may correspond to the wrong instance among small and repeated objects. We call this mismatch between mask rendering and expression-level target selection the Grounding-Rendering Gap. We present DroneSegEdit, which combines RGB proposal rendering with language-grounded, proposal-constrained verification to select supported foreground while retaining rendered geometry. DroneSegEdit adapts the open-weight LongCat-Image-Edit model using task-specific LoRA and fixed color-to-label decoding. For referring segmentation, a Language-Grounded Spatial Verifier (LGSV) reads hierarchical states from the frozen multimodal large language model backbone to derive expression-conditioned spatial evidence, which is mapped back to the rendered proposal without introducing foreground outside it. On RIS-LAD, DroneSegEdit reaches a new state of the art at 52.78 mIoU, 7.16 points above the previous best. It remains competitive overall on UAVid-RIS and VDD-RIS, with its clearest class-wise gains on compact, application-relevant categories, notably vehicles and humans. A separate verifier-free experiment extends the RGB interface to full-scene parsing: a task-specific adapter jointly renders all classes from one task-level instruction without enumerating class names or palette colors.

cs.CV↗

Friendship is All we Need: A Multi-graph Embedding Approach for Modeling Customer Behavior

Understanding customer behavior is fundamental for many use-cases in industry, especially in accelerated growth areas such as fin-tech and e-commerce. Structured data are often expensive, time-consuming and inadequate to analyze and study complex customer behaviors. In this paper, we propose a multi-graph embedding approach for creating a non-linear representation of customers in order to have a better knowledge of their characteristics without having any prior information about their financial status or their interests. By applying the current method we are able to predict users' future behavior with a reasonably high accuracy only by having the information of their friendship network. Potential applications include recommendation systems and credit risk forecasting.

cs.LG↗

Modeling Pharmacological Effects with Multi-Relation Unsupervised Graph Embedding

A pharmacological effect of a drug on cells, organs and systems refers to the specific biochemical interaction produced by a drug substance, which is called its mechanism of action. Drug repositioning (or drug repurposing) is a fundamental problem for the identification of new opportunities for the use of already approved or failed drugs. In this paper, we present a method based on a multi-relation unsupervised graph embedding model that learns latent representations for drugs and diseases so that the distance between these representations reveals repositioning opportunities. Once representations for drugs and diseases are obtained we learn the likelihood of new links (that is, new indications) between drugs and diseases. Known drug indications are used for learning a model that predicts potential indications. Compared with existing unsupervised graph embedding methods our method shows superior prediction performance in terms of area under the ROC curve, and we present examples of repositioning opportunities found on recent biomedical literature that were also predicted by our method.

cs.LG↗