Oracle-Robust Online Alignment for Large Language Models
cs.LG, stat.ML
Submitted: 2026-02-24
Updated: 2026-11-04
Terminology
Sources
- Fine-Tuning Language Models from Human Preferences
- Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
- Robust LLM Alignment via Distributionally Robust Direct Preference Optimization
- Goodhart's Law in Reinforcement Learning
- Helping or Herding? Reward Model Ensembles Mitigate but do not Eliminate Reward Hacking
- SAIL: Self-Improving Efficient Online Alignment of Large Language Models
- Online Preference Alignment for Language Models via Count-based Exploration
- Self-Exploring Language Models: Active Preference Elicitation for Online Alignment
- Towards Robust Alignment of Language Models: Distributionally Robustifying Direct Preference Optimization
- PARL: A Unified Framework for Policy Alignment in Reinforcement Learning from Human Feedback
- On The Sample Complexity Bounds In Bilevel Reinforcement Learning
- Proximal Policy Optimization Algorithms
- Right Now, Wrong Then: Non-Stationary Direct Preference Optimization under Preference Drift
- Distributionally Robust Optimization: A Review
- Stochastic subgradient method converges at the rate $O(k^{-1/4})$ on weakly convex functions
Related papers
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis
- (Mis)generalization of Helpful-Only Fine-tuning