CU-DPO extends DPO with continuous utility scores for strategy portfolios, proving Theta(K log K) better sample complexity and reporting 68-78% strategy selection accuracy plus up to 6.6 point reasoning gains.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Continuous-Utility Direct Preference Optimization
CU-DPO extends DPO with continuous utility scores for strategy portfolios, proving Theta(K log K) better sample complexity and reporting 68-78% strategy selection accuracy plus up to 6.6 point reasoning gains.