Enforcing Axioms for AI Alignment under Loss-Based Rules
cs.GT, cs.AI, cs.LG
Submitted: 2026-11-09
Updated: 2026-11-09
Terminology
Sources
- Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
- Constitutional AI: Harmlessness from AI Feedback
- Preference Restrictions in Computational Social Choice: A Survey
- Improving alignment of dialogue agents via targeted human judgements
- Distortion of AI Alignment: Does Preference Optimization Optimize for Preferences?
- Beyond RLHF and NLHF: Population-Proportional Alignment under an Axiomatic Framework
- Clone-Robust AI Alignment
- Representative Social Choice: From Learning Theory to AI Alignment
- Theoretical Tensions in RLHF: Reconciling Empirical Success with Inconsistencies in Social Choice Theory
Related papers
- How Many Votes is a Lie Worth? Measuring Strategyproofness through Resource Augmentation
- Effective Horizons for Prophet Inequalities with Discounted I.I.D. Rewards
- Truthful-in-Expectation Mechanism with Constant Maximin-Share Guarantee
- The Power of Recruiting the Smaller Side: Two Additional Traders Suffice in Two-Sided Markets
- From signaling to interviews in random matching markets
- Optimizing Social Utility in Sequential Experiments