EvoPref applies NSGA-II evolutionary optimization with archive-based diversity to populations of LoRA adapters, yielding 18% higher preference coverage and 47% lower collapse than gradient descent baselines while matching alignment quality.
Montes de Oca, Thomas Stützle, and Marco Dorigo
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5representative citing papers
OpenEAI-Platform delivers an open-source low-cost robotic arm and VLA model that outperforms commercial arms and matches large pretrained baselines on four real-world manipulation tasks using limited open data.
A multi-objective probabilistic forecast combination framework is introduced that generates Pareto-optimal combinations balancing forecast accuracy and inventory decision performance, outperforming single-objective methods on retail and spare parts data.
MERSEM uses evolutionary reinforcement learning to allocate graph workloads in edge-cloud systems, reducing SLA violations by up to 45% and carbon emissions by up to 12% versus prior methods.
FAmv modifies the Firefly Algorithm with a unified hybrid distance for mixed continuous-discrete spaces and matches or exceeds state-of-the-art methods on CEC2013 benchmarks and engineering problems.
citing papers explorer
-
EvoPref: Multi-Objective Evolutionary Optimization Discovers Diverse LLM Alignments Beyond Gradient Descent
EvoPref applies NSGA-II evolutionary optimization with archive-based diversity to populations of LoRA adapters, yielding 18% higher preference coverage and 47% lower collapse than gradient descent baselines while matching alignment quality.
-
OpenEAI-Platform: An Open-source Embodied Artificial Intelligence Hardware-Software Unified Platform
OpenEAI-Platform delivers an open-source low-cost robotic arm and VLA model that outperforms commercial arms and matches large pretrained baselines on four real-world manipulation tasks using limited open data.
-
Multi-objective probabilistic forecast combination for inventory demand
A multi-objective probabilistic forecast combination framework is introduced that generates Pareto-optimal combinations balancing forecast accuracy and inventory decision performance, outperforming single-objective methods on retail and spare parts data.
-
Sustainable Graph Analytics Workload Scheduling with Evolutionary Reinforcement Learning in Edge-Cloud Systems
MERSEM uses evolutionary reinforcement learning to allocate graph workloads in edge-cloud systems, reducing SLA violations by up to 45% and carbon emissions by up to 12% versus prior methods.
-
A Firefly Algorithm for Mixed-Variable Optimization Based on Hybrid Distance Modeling
FAmv modifies the Firefly Algorithm with a unified hybrid distance for mixed continuous-discrete spaces and matches or exceeds state-of-the-art methods on CEC2013 benchmarks and engineering problems.