5 papers
Health-ORSC-Bench: A Benchmark for Measuring Over-Refusal and Safety Completion in Health Context
Zhihao Zhang, Liting Huang, Guanghao Wu +3
Safety alignment in Large Language Models is critical for healthcare; however, reliance on binary refusal boundaries often results in over-refusal of benign queries or unsafe compl…
Divergent Paths to Depolarization: Dialogue Design Shapes the Intergroup Attitudinal Effects of AI-Assisted Political Argumentation
Jianlong Zhu, Syed Muhammad Jhon Raza Naqvi, Carolin-Theresa Ziemer +2
Structured argumentative dialogues where interlocutors deliberate on opposing political ideas are known to promote perspective-taking and reduce political polarization, but finding…
CBT-Audio: Evaluating Audio Language Models for Patient-Side Distress Intensity Estimation in CBT Session Recordings
Qixuan Hu, Shuchang Ye, Xumou Zhang +8
Cognitive behavioural therapy is widely used to help patients understand and manage psychological distress. It is often delivered through spoken conversation, where therapists atte…
RadGenome-Anatomy: A Large-Scale Anatomy-Labeled Chest Radiograph Dataset via Physically Grounded Volumetric Projection
Shuchang Ye, Mingyuan Meng, Hao Wang +2
Anatomical structure labels for chest radiographs are essential for medical image segmentation and a broad range of downstream diagnostic tasks. However, annotating anatomy directl…
VISPA: Pluralistic Alignment via Automatic Value Selection and Activation
Shenyan Zheng, Jiayou Zhong, Anudeex Shetty +3
As large language models are increasingly used in high-stakes domains, it is essential that their outputs reflect not average} human preference, rather range of varying perspective…