Which Preferences to Train On? End-to-End Multi-Objective Alignment with an Adversarial Preference DistributionMinjae Lee,Kyunghyun Cho, Sangdon Park2026Last updated on Oct 4, 2026 ← Stopping Is Not the Limit: Understanding and Eliciting Agent Capability in Long-Horizon Vulnerability Exploitation Nov 17, 2026Online Conformal Abstention Under Adversarial Bandit Feedback Sep 26, 2026 →