6 papers
FlipAttack: Jailbreak LLMs via Flipping
Yue Liu, Xiaoxin He, Miao Xiong +5
This paper proposes a simple yet effective jailbreak attack named FlipAttack against black-box LLMs. First, from the autoregressive nature, we reveal that LLMs tend to understand t…
KLong: Training LLM Agent for Extremely Long-horizon Tasks
Yue Liu
This paper introduces KLong, an open-source LLM agent trained to solve extremely long-horizon tasks. The principle is to first cold-start the model via trajectory-splitting SFT, th…
EvoTest: Evolutionary Test-Time Learning for Self-Improving Agentic Systems
Yufei He, Juncheng Liu, Yue Liu +5
A fundamental limitation of current AI agents is their inability to learn complex skills on the fly at test time, often behaving like "clever but clueless interns" in novel environ…
Towards Realistic Personalization: Evaluating Long-Horizon Preference Following in Personalized User-LLM Interactions
Qianyun Guo, Yibo Li, Yue Liu +1
Large Language Models (LLMs) are increasingly serving as personal assistants, where users share complex and diverse preferences over extended interactions. However, assessing how w…
VPI-Bench: Visual Prompt Injection Attacks for Computer-Use Agents
Tri Cao, Bennett Lim, Yue Liu +7
Computer-Use Agents (CUAs) with full system access enable powerful task automation but pose significant security and privacy risks due to their ability to manipulate files, access…
UniGraph2: Learning a Unified Embedding Space to Bind Multimodal Graphs
Yufei He, Yuan Sui, Xiaoxin He +3
Existing foundation models, such as CLIP, aim to learn a unified embedding space for multimodal data, enabling a wide range of downstream web-based applications like search, recomm…