Skip to content
NeurIPS 2026 planner

Poster · Sydney

AERO: Adaptive Ensemble-Disagreement Routing for Oracle Feedback in Sample-Efficient Online RLHF

When
Tue 8 Dec 2026, 10:00–13:00 AEDT
Where
Hall 1-4, ICC Sydney
Exhibition Centre, Level 1
Session
Sydney Poster Session 1
Authors
Shrey Singhal, Nannan Huang, Haytham Fayek
Organisations
RMIT University, RMIT
QRS classification
Reinforcement learning, Language models

Open in scheduleOfficial page