一作ACL 2026Poster· Natural Language Processing
Where Paths Split: Localized, Calibrated Control of Moral Reasoning in Large Language Models
Chenchen Yuan · Zheyu Zhang · Gjergji Kasneci
摘要
Large language models often display heterogeneous moral preferences across settings. We study inference-time steering toward a desired ethical framework while preserving general competence. We present Convergent-Divergent Routing, which traces and edits minimal branch points inside transformer bl...