Publications (7)
Strategic Fusion of Vision Language Models: Shapley-Credited Context-Aware Dawid-Skene for Multi-Label Tasks in Autonomous Driving
Yuxiang Feng, Keyang Zhang, Hassane Ouchouid +3
Large vision-language models (VLMs) are increasingly used in autonomous-vehicle (AV) stacks, but hallucination limits their reliability in safety-critical pipelines. We present Sha…
Learning from Mixed Datasets: A Monotonic Image Quality Assessment Model
Zhaopeng Feng, Keyang Zhang, Shuyue Jia +2
Deep learning based image quality assessment (IQA) models usually learn to predict image quality from a single dataset, leading the model to overfit specific scenes. To account for…
Propose and Rectify: A Forensics-Driven MLLM Framework for Image Manipulation Localization
Keyang Zhang, Chenqi Kong, Hui Liu +3
The increasing sophistication of image manipulation techniques demands robust forensic solutions that can both reliably detect alterations and precisely localize tampered regions.…
LAST: The Last Query Token Guides Visual Token Pruning for Edge-Cloud Collaborative MLLM Inference
Feng Yang, Xinrui Ju, Keyang Zhang +6
The paper introduces LAST, a training‑free method that uses the attention of the last query token to prune visual tokens on edge devices before sending them to a cloud multimodal L…
Lightweight Yet Secure: Secure Scripting Language Generation via Lightweight LLMs
Keyang Zhang, Zeyu Chen, Xuan Feng +4
The security of scripting languages such as PowerShell is critical given their powerful automation and administration capabilities, often exercised with elevated privileges. Today,…
Image Provenance Analysis via Graph Encoding with Vision Transformer
Keyang Zhang, Chenqi Kong, Shiqi Wang +2
Recent advances in AI-powered image editing tools have significantly lowered the barrier to image modification, raising pressing security concerns those related to spreading misinf…
Convergence of the boundary integral method for interfacial Stokes flow
David M. Ambrose, Michael Siegel, Keyang Zhang
Boundary integral numerical methods are among the most accurate methods for interfacial Stokes flow, and are widely applied. They have the advantage that only the boundary of the d…