8 papers
Convergent Detour Hijacking: Task-Preserving Resource Amplification in Skill-Based LLM Agents
Junliang Liu, Ruoyu Li, Wenxin Tang +4
LLM agents increasingly rely on third-party skills, using natural-language descriptions for selection and instruction bodies for planning. This progressive-disclosure design expose…
CodeShrink: Adaptive Visual Compression for Efficient Multimodal Code Understanding
Wenxin Tang, Jingyu Xiao, Zhenyu Liu +6
Rendering source code as images offers a promising way to reduce the input costs of Multimodal Large Language Models (MLLMs). Adjusting image resolution can trade visual token cost…
VulTriage: Triple-Path Context Augmentation for LLM-Based Vulnerability Detection
Wenxin Tang, Xiang Zhang, Junliang Liu +11
Automated vulnerability detection is a fundamental task in software security, yet existing learning-based methods still struggle to capture the structural dependencies, domain-spec…
DCVD: Dual-Channel Cross-Modal Fusion for Joint Vulnerability Detection and Localization
Wenxin Tang, Wenbin Li, Junliang Liu +10
Software vulnerability detection plays a critical role in ensuring system security, where real-world auditing requires not only determining whether a function is vulnerable but als…
DesignBench: A Comprehensive Benchmark for MLLM-based Front-end Code Generation
Jingyu Xiao, Ming Wang, Man Ho Lam +4
Multimodal Large Language Models (MLLMs) have demonstrated remarkable capabilities in automated front-end engineering, e.g., generating UI code from visual designs. However, existi…
Benchmarking MLLM-based Web Understanding: Reasoning, Robustness and Safety
Junliang Liu, Jingyu Xiao, Wenxin Tang +5
Multimodal large language models (MLLMs) are increasingly deployed as the core reasoning engine for web-facing systems, powering GUI agents and front-end automation that must inter…