Revisiting scaling laws for reward optimization

arXiv:2609.38526 · cs.LG, cs.AI · Submitted 2026-09-29 · Read on arXiv

cs.LG, cs.AI

Submitted: 2026-09-29

Updated: 2026-09-29

Terminology

• Scaling Laws — 1× in this paper · explained in 4 episodes

Sources

• Best-of-N through the Smoothing Lens: KL Divergence and Regret Analysis
• Helping or Herding? Reward Model Ensembles Mitigate but do not Eliminate Reward Hacking
• AI Safety Gridworlds
• Categorizing Variants of Goodhart's Law
• Consequentialist Objectives and Catastrophe
• Information Theoretic Guarantees For Policy Alignment In Large Language Models
• WebGPT: Browser-assisted question-answering with human feedback
• Soft Best-of-n Sampling for Model Alignment

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: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks