4 papers
VIABLE: A Visually Impaired Assistance Benchmark for VLM-as-a-Judge Evaluation
Yi Zhao, Siqi Wang, Zhe Hu +2
AI-based Visually Impaired Assistance (VIA) remains challenging, largely due to the high cost of human evaluation. The VLM-as-a-Judge paradigm may offer a promising alternative, al…
CitySeeker: How Do VLMS Explore Embodied Urban Navigation With Implicit Human Needs?
Siqi Wang, Chao Liang, Yunfan Gao +5
Vision-Language Models (VLMs) have made significant progress in explicit instruction-based navigation; however, their ability to interpret implicit human needs (e.g., "I am thirsty…
Sighted by Default: Addressing Implicit Vision Assumptions in Real-Time VLM Assistance for BLV Users
Yi Zhao, Siqi Wang, Qiqun Geng +2
Vision-Language Model (VLM)-based assistance is reshaping independence for blind and low-vision (BLV) users, yet current tools fail in dynamic settings. While request-response arch…
CoSafe: Evaluating Large Language Model Safety in Multi-Turn Dialogue Coreference
Erxin Yu, Jing Li, Ming Liao +4
As large language models (LLMs) constantly evolve, ensuring their safety remains a critical research problem. Previous red-teaming approaches for LLM safety have primarily focused…