The paper introduces MOMEHA and MB-MOMEHA, Hessian-free single-loop algorithms that converge to relaxed Pareto-stationary points for multi-objective bilevel problems with nonconvex lower levels.
Pareto low-rank adapters: Efficient multi-task learning with preferences
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Efficient Hessian-Free Methods for Multi-Objective Bilevel Optimization with Nonconvex Lower Level
The paper introduces MOMEHA and MB-MOMEHA, Hessian-free single-loop algorithms that converge to relaxed Pareto-stationary points for multi-objective bilevel problems with nonconvex lower levels.