AI RESEARCH
Stance Labels Fail When They Matter Most: The Projection Problem in Stance Detection
arXiv CS.CL
•
ArXi:2603.24231v1 Announce Type: new Stance detection is nearly always formulated as classifying text into Favor, Against, or Neutral -- a convention inherited from debate analysis and applied without modification to social media since SemEval-2016. But attitudes toward complex targets are not unitary: a person can accept climate science while opposing carbon taxes, expressing on one dimension and opposition on another.