Human-Residual Steering: Aligning Frozen LLM Judges with Human Judgment Distributions
arXiv:2610.37868v1 Announce Type: new Abstract: Large language models (LLMs) are widely used as automated judges, yet most alignment methods target aggregated human judgments and overlook variation among individual annotators. Existing approaches to matching the full distribution of human ratings ma
Der Artikeltext wird beim nächsten Sync nachgeladen. Original bei arXiv cs.CL lesen
Quelle: arXiv cs.CL (arXiv cs.CL). Original öffnen