Sterling Research Group

Cross model transfer of Direct Preference Optimization

Since direct preference optimization (DPO) was introduced in 2023, it has become a widely used preference optimization strategy for post-training large language models (LLMs). DPO's implicit reward formulation is coupled to its reference policy, and its theoretical guarantees ass...

Cross model transfer of Direct Preference Optimization

Since direct preference optimization (DPO) was introduced in 2023, it has become a widely used preference optimization strategy for post-training large language models (LLMs). DPO's implicit reward formulation is coupled to its reference policy, and its theoretical guarantees ass...

Cross model Direct Preference Optimization

Since direct preference optimization (DPO) was introduced in 2023, it has become a widely used preference optimization strategy for post-training large language models (LLMs). DPO's implicit reward formulation is coupled to its reference policy, and its theoretical guarantees ass...

Graded Branch Unpredictability Near Nonlinear Transitions: A Decision-Scale Framework and Held-Out Energy-Balance Demonstration

Forecast laws near nonlinear transitions may be poorly concentrated even when they are not cleanly bimodal. For a conditional law P(t,h) on a metric state space, we study Uδ(t,h) = 1 − supₐ P(t,h){d(Y,a) ≤ δ}: the minimum probability that a point forecast misses the realization b...