RMB: Reward Model Boosting Mitigates Reward Hacking
cs.LG
Submitted: 2026-09-27
Updated: 2026-09-27
Code: https://github.com/MANGA-UOFA/RMB
Terminology
Sources
- A General Language Assistant as a Laboratory for Alignment
- Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Gemma: Open Models Based on Gemini Research and Technology
- ROPO: Robust Preference Optimization for Large Language Models
- Llama 2: Open Foundation and Fine-Tuned Chat Models
- GPT-4 Technical Report
- Qwen2.5 Technical Report
- WARM: On the Benefits of Weight Averaged Reward Models
- Proximal Policy Optimization Algorithms
- Secrets of RLHF in Large Language Models Part II: Reward Modeling
- Reward Model Overoptimisation in Iterated RLHF
- reWordBench: Benchmarking and Improving the Robustness of Reward Models with Transformed Inputs
- Reward-Robust RLHF in LLMs
- Uncertainty-Penalized Reinforcement Learning from Human Feedback with Diverse Reward LoRA Ensembles
- Overcoming Reward Overoptimization via Adversarial Policy Optimization with Lightweight Uncertainty Estimation
- Iterative Data Smoothing: Mitigating Reward Overfitting and Overoptimization in RLHF
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks