Ke Xu

A look into the safety concerns surrounding large language models.

Exploring risks of model inversion attacks on private data in machine learning models.

This study investigates the role of radio-excess AGNs in galaxy evolution.

A new method enhances shadow detection accuracy in low-light areas.

A new system enhances detection of Internet routing anomalies using BGP semantics.

This framework allows collaborative model training while ensuring data privacy and model integrity.

Vimentin plays a vital role in maintaining cell shape and responding to environmental changes.

Explore essential techniques for handling large datasets effectively.

JWST reveals unique patterns in star formation within galaxy cluster CLJ1001.

HILL improves hierarchical text classification by preserving data integrity and structure.

A new method enhances AI's ability to edit knowledge and answer complex questions.

A method to forecast non-factual answers from language models before they generate responses.

New method reveals vulnerabilities in Vision-Language Pre-training models through universal adversarial perturbations.

Examine various jailbreak attacks on language models and their defenses.

Introducing a method to automate 3D labeling using 2D prompts.

A novel approach to recognize categories in unlabeled data while preserving old classifications.

New actuator designs enhance efficiency for underwater microrobots.

Learn how to detect and combat rumors online effectively.

3C 59 revives star formation after a long slump, surprising scientists.

Engorgio exploits language models, raising concerns about service reliability and performance.