Showing 2026Show all
2 papers · 1 filter
cs.LG2026
Uncertainty-Normalized Margins for Direct Preference Optimization
Sadegh Khorasani, Petrus Mikkola, Matthias Grossglauser
Direct preference optimization (DPO) models binary preferences through a Bradley-Terry model with a common noise scale, without explicitly accounting for preference strength or pro…
cs.AI2026
Inference Time Causal Probing in LLMs
Sadegh Khorasani, Saber Salehkaleybar, Negar Kiyavash +1
Causal probing methods aim to test and control how internal representations influence the behavior of generative models. In causal probing, an intervention modifies hidden states s…