Jacky Keung

HumanEval-V: Systematic Evaluation of Visual Reasoning in Large Multimodal Models for Code Generation

Fengji Zhang, Linquan Wu, Huiyu BAI, Guancheng Lin et autres

Visual representations are essential for communicating complex programming concepts and play a critical role in algorithm and code design. While recent advances in Large Multimodal Models (LMMs) have demonstrated impressive visual capabilities across various domains, their visual reasoning abilities in coding contexts …

hk, cn (code pays fourni par la source)

KBC-Coop: Knowledge-Grounded Role Cooperation for Single-Model LLMs in Collaborative Tasks

H Q Yang, Linquan Wu, Shaochao Lin, Zixuan Wang et autres

cn, hk (code pays fourni par la source)

Chart2Code-MoLA: Efficient Multi-Modal Code Generation via Adaptive Expert Routing

Yifei Wang, Jacky Keung, Zhenyu Mao, Jingyu Zhang et autres

Chart-to-code generation is a critical task in automated data visualization, translating complex chart structures into executable programs. While recent Multi-modal Large Language Models (MLLMs) improve chart representation, existing approaches still struggle to achieve cross-type generalization, memory efficiency, and modular design. To address …

hk (code pays fourni par la source)

PerProb: Indirectly Evaluating Memorization in Large Language Models

Yihan Liao, Jacky Keung, Xiaoxue Ma, Jingyu Zhang et autres

The rapid advancement of Large Language Models (LLMs) has been driven by extensive datasets that may contain sensitive information, raising serious privacy concerns. One notable threat is the Membership Inference Attack (MIA), where adversaries infer whether a specific sample was used in …

hk, bd (code pays fourni par la source)

Understanding Industrial Log Analysis: A Multi-Dataset Evaluation of Parsing and Anomaly Detection

Yicheng Sun, Jacky Keung, Yihan Liao, Hi Kuen Yu

Log analysis plays a critical role in monitoring and maintaining the safety of industrial software systems. However, most existing research relies heavily on benchmark datasets derived from legacy or open-source systems, which fail to capture the structural diversity and operational complexity of …

hk (code pays fourni par la source)

Towards Lightweight LLM Software Solutions for InsurTech: A Framework for Scalable Question Answering Systems

Hi Kuen Yu, Jacky Keung, Yicheng Sun, Yihan Liao et autres

The integration of Large Language Models (LLMs) into software systems is transforming regulated sectors like insurance, where precision, compliance, and efficiency are essential. While proprietary LLMs like GPT-4 offer state-of-the-art performance, their closed-source nature and high computational demands constrain adoption in privacy-sensitive …

hk, cn (code pays fourni par la source)