8 papers
Devil in the Lens: Analyzing and Defending Physical Prompt Injection Against Vision-Language Models on Wearable Devices
Yaxin Li, Hao Wang, Yanda Shao +2
Vision-Language Models (VLMs) are rapidly deployed on human-facing wearable devices such as smart glasses to enable multimodal perception and AI-assisted decision-making. While pri…
Send a SCOUT First: Pre-hoc Reasoning for Adaptive Detector Allocation in Prompt-Injection Defense
Shuhao Zhang, Jiarui Li, Qi Cao +2
Prompt-injection detectors are heterogeneous: each is strong on a different slice of attacks, and none is always reliable. Yet existing systems still treat detection as a fixed sin…
Mental Damage: Caption Poisoning Attacks on Retrieval-Augmented Text-to-Music Generation
Yizhu Wen, Shuhao Zhang, Nan Zhang +2
Retrieval-augmented text-to-music (TTM) systems augment underspecified user prompts using captions retrieved from a music caption dataset. This design introduces an integrity depen…
LLMs Know When They Know, but Do Not Act on It: A Metacognitive Harness for Test-time Scaling
Qi Cao, Yufan Wang, Peijia Qin +2
Large language models (LLMs) often expose useful signals of self-monitoring: before solving a problem, they can estimate whether they are likely to succeed, and after solving it, t…
PruneTIR: Inference-Time Tool Call Pruning for Effective yet Efficient Tool-Integrated Reasoning
Luan Zhang, Dandan Song, Zhijing Wu +8
Tool-integrated reasoning (TIR) enables large language models (LLMs) to enhance their capabilities by interacting with external tools, such as code interpreters (CI). Most recent s…
ActiShade: Activating Overshadowed Knowledge to Guide Multi-Hop Reasoning in Large Language Models
Huipeng Ma, Luan Zhang, Dandan Song +10
In multi-hop reasoning, multi-round retrieval-augmented generation (RAG) methods typically rely on LLM-generated content as the retrieval query. However, these approaches are inher…