Binarization Flattens the Score Space
cs.LG, cs.AI
Submitted: 2026-09-17
Updated: 2026-09-17
Sources
- Concrete Problems in AI Safety
- HealthBench: Evaluating Large Language Models Towards Improved Human Health
- Scaling Laws for Reward Model Overoptimization
- Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains
- Measuring Mathematical Problem Solving With the MATH Dataset
- Compute as Teacher: Turning Inference Compute Into Reference-Free Supervision
- Goodhart's Law in Reinforcement Learning
- Nine Judges, Two Effective Votes: Correlated Errors Undermine LLM Evaluation Panels
- Let's Verify Step by Step
- Reward Hacking in Rubric-Based Reinforcement Learning
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Judging the Judges: A Systematic Study of Position Bias in LLM-as-a-Judge
- Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models
- Reproducing, Analyzing, and Detecting Reward Hacking in Rubric-Based Reinforcement Learning
- Rubric Dropout: A Simple Way to Mitigate Reward Hacking in Rubric-as-Reward RL
- Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena
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