172 citations · 610 across the 14 of their papers we have counts for
Showing cs.CLShow all
2 papers · 1 filter
cs.CL2026
Beyond the Sycophancy Score: How Task, Model, and Pressure Shape LLM Yielding
Guang Yang, Homa Hosseinmardi, Fengchen Liu +1
Large language models (LLMs) often abandon a correct answer, or endorse a user's position, once the user pushes back. This behavior, called sycophancy, is usually reported as a sin…
cs.CL2026
Beyond Binary Detection: A Multi-Dimensional Taxonomy of Cancer Misinformation on Reddit
Aria Pessianzadeh, Pooriya Jamie, Naima Sultana +6
Cancer-related discussions on social media provide important spaces for information exchange and peer support, but can also expose users to misinformation with implications for pre…