MAdam preconditions MOO solver directions with preference-conditioned curvature so that Adam's adaptive steps respect the intended metric instead of entangling it with gradient history.
arXiv preprint arXiv:2308.12029 , year =
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
MI-EPO maximizes joint conditional mutual information among responses, feedback, and preference vectors, using probabilistic routing to improve alignment and controllability in multi-objective LLM optimization.
citing papers explorer
-
MAdam: Metric-Aware Multi-Objective Adam
MAdam preconditions MOO solver directions with preference-conditioned curvature so that Adam's adaptive steps respect the intended metric instead of entangling it with gradient history.
-
Multi-Objective Exploration and Preference Optimization via Mutual Information
MI-EPO maximizes joint conditional mutual information among responses, feedback, and preference vectors, using probabilistic routing to improve alignment and controllability in multi-objective LLM optimization.