8 papers
BAIT: Boundary-Guided Disclosure Escalation via Self-Conditioned Reasoning
Xuan Luo, Yue Wang, Geng Tu +2
In this work, we propose BAIT (Boundary-Aware Iterative Trap), a three-step jailbreak framework that approaches malicious goals through internal disclosure. BAIT first asks the mod…
Mitigating Provenance-Role Collapse in Long-Term Agents via Typed Memory Representation
Zhengda Jin, Bingbing Wang, Jing Li +2
Long-term memory is essential for persistent LLM agents, yet prevailing architectures store historical interactions as unstructured, flat text. This unconstrained storage induces p…
Proactive Memory for Ad-Hoc Recall over Streaming Dialogues
Bingbing Wang, Jing Li, Ruifeng Xu
Real-world dialogue usually unfolds as an infinite stream. It thus requires bounded-state memory mechanisms to operate within an infinite horizon. However, existing read-then-think…
Speech-Omni-Lite: Portable Speech Interfaces for Vision-Language Models
Dehua Tao, Xuan Luo, Daxin Tan +5
While large-scale omni-models have demonstrated impressive capabilities across various modalities, their strong performance heavily relies on massive multimodal data and incurs sub…
PRISM of Opinions: A Persona-Reasoned Multimodal Framework for User-centric Conversational Stance Detection
Bingbing Wang, Zhixin Bai, Zhengda Jin +6
The rapid proliferation of multimodal social media content has driven research in Multimodal Conversational Stance Detection (MCSD), which aims to interpret users' attitudes toward…
A Simple and Efficient Jailbreak Method Exploiting LLMs' Helpfulness
Xuan Luo, Yue Wang, Zefeng He +3
This study reveals a critical safety blind spot in modern LLMs: learning-style queries, which closely resemble ordinary educational questions, can reliably elicit harmful responses…