PEARL: Prediction Error-Adaptive Regularized Learning for Selective Forgetting under Partial Preference Drift
cs.LG, cs.AI
Submitted: 2026-11-04
Updated: 2026-11-04
Terminology
Sources
- Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
- One Adapts to Any: Meta Reward Modeling for Personalized LLM Alignment
- Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference
- UltraFeedback: Boosting Language Models with Scaled AI Feedback
- Uncertainty-Aware Variational Reward Factorization via Probabilistic Preference Bases for LLM Personalization
- Uncertainty-aware Reward Model: Teaching Reward Models to Know What is Unknown
- T-POP: Test-Time Personalization with Online Preference Feedback
- Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks
- A Roadmap to Pluralistic Alignment
- Personalized Group Relative Policy Optimization for Heterogenous Preference Alignment
- Uncertainty-Penalized Reinforcement Learning from Human Feedback with Diverse Reward LoRA Ensembles
- Neural Thompson Sampling
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- 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