14 papers
Measuring the Security of Mobile LLM Agents under Adversarial Prompts from Untrusted Third-Party Channels
Chenghao Du, Quanfeng Huang, Tingxuan Tang +3
Large Language Models (LLMs) have transformed software development, enabling AI-powered applications known as LLM-based agents that promise to automate tasks across diverse apps an…
Hierarchical Concept-to-Appearance Guidance for Multi-Subject Image Generation
Yijia Xu, Zihao Wang, Haokun Gui +1
Multi-subject image generation aims to synthesize images that faithfully preserve the identities of multiple reference subjects while following textual instructions. However, exist…
Knowledge-Graph-Driven Data Synthesis for Low-Resource Software Development: A HarmonyOS Case Study
Mingwei Liu, Zheng Pei, Yanlin Wang +5
In low-resource framework development (e.g., HarmonyOS), large language models (LLMs) often lack sufficient pre-training exposure, resulting in poor code generation performance. Al…
DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
DeepSeek-AI, Anyi Xu, Bangcai Lin +315
We present a preview version of DeepSeek-V4 series, including two strong Mixture-of-Experts (MoE) language models -- DeepSeek-V4-Pro with 1.6T parameters (49B activated) and DeepSe…
Activation-Guided Local Editing for Jailbreaking Attacks
Jiecong Wang, Haoran Li, Hao Peng +4
Jailbreaking is an essential adversarial technique for red-teaming these models to uncover and patch security flaws. However, existing jailbreak methods face significant drawbacks.…
A User-Friendly Framework for Generating Model-Preferred Prompts in Text-to-Image Synthesis
Nailei Hei, Qianyu Guo, Zihao Wang +3
Well-designed prompts have demonstrated the potential to guide text-to-image models in generating amazing images. Although existing prompt engineering methods can provide high-leve…