Zhang Fan
8 indexed papers
Publications per year
Top categories
Frequent co-authors
Research Timeline
This paper provides the first comprehensive systematization and large-scale empirical evaluation of existing LLM-based Automated Penetration Testing (AutoPT) frameworks, offering a structured taxonomy and unified benchmark for the field.
This paper introduces a novel, efficient protocol for training Gradient Boosting Decision Trees (GBDT) on vertically partitioned data held by two mutually distrustful parties while ensuring complete anonymity by hiding record identifiers.
The paper introduces CardioLens, a rigorous evaluation testbed for multi-sequence Cardiac MRI, which reveals that current Multimodal Large Language Models (MLLMs) exhibit a significant 'clinical reality gap' and perform poorly when simulating real-world cardiac interpretation workflows.
The paper introduces CardioLens, a rigorous evaluation testbed for multi-sequence Cardiac MRI, which reveals that current Multimodal Large Language Models (MLLMs) exhibit a significant 'clinical reality gap' and perform poorly when simulating real-world cardiac interpretation workflows.
The paper proposes FedMChain, a novel federated learning framework that structures multimodal training into sequential phases to mitigate modality competition and improve model performance while reducing communication overhead.
The paper proposes FedMChain, a novel federated learning framework that structures multimodal training into sequential phases to mitigate modality competition and improve model performance while reducing communication overhead.
This paper investigates and quantifies API key leakage in iOS applications that use large language models.
Argus is a persistent, self-evolving runtime system that achieves better long-horizon reasoning results than Direct Copilot in seven GPT-5.5 benchmark arenas, with improvements in solve-input tokens, active workflow time, and verification-gated self-evolution.
Papers
Argus: A General-Purpose Agentic Runtime for Long-Horizon Reasoning
Boxiu Li, Zimo Wen, Yijia Fan, Junxiang Lei +22 more
Argus is a persistent, self-evolving runtime system that achieves better long-horizon reasoning results than Direct Copilot in seven GPT-5.5 benchmark arenas, with improvements in solve-input tokens,…