4 papers
Correctness-Optimized Residual Activation Lens (CORAL): Transferrable and Calibration-Aware Inference-Time Steering
Miranda Muqing Miao, Young-Min Cho, Lyle Ungar
Large language models (LLMs) exhibit persistent miscalibration, especially after instruction tuning and preference alignment. Modified training objectives can improve calibration,…
Designing Mental-Health Chatbots for Indian Adolescents: Mixed-Methods Evidence, a Boundary-Object Lens, and a Design-Tensions Framework
Neil K. R. Sehgal, Hita Kambhamettu, Sai Preethi Matam +2
Mental health challenges among Indian adolescents are shaped by unique cultural and systemic barriers, including high social stigma and limited professional support. We report a mi…
Towards Style Alignment in Cross-Cultural Translation
Shreya Havaldar, Adam Stein, Eric Wong +1
Successful communication depends on the speaker's intended style (i.e., what the speaker is trying to convey) aligning with the listener's interpreted style (i.e., what the listene…
Large Language Models Show Human-like Social Desirability Biases in Survey Responses
Aadesh Salecha, Molly E. Ireland, Shashanka Subrahmanya +3
As Large Language Models (LLMs) become widely used to model and simulate human behavior, understanding their biases becomes critical. We developed an experimental framework using B…