Beyond Preferences in AI Alignment

Iterative Reasoning Preference Optimization

Humans prefer AI-generated copy, survey finds

Self-Rewarding Language Models

Diffusion Model Alignment Using Direct Preference Optimization

The Alignment Problem: Machine Learning and Human Values with Brian Christian

Google Bard readies ‘Memory’ to adapt to important details about you

The Jiminy Advisor: Moral Agreements among Stakeholders Based on Norms and Argumentation | Journal of Artificial Intelligence Research

Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesis

Emotion prediction as computation over a generative theory of mind | Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences

Direct Preference Optimization: Your Language Model is Secretly a Reward Model

Pick-a-Pic: An Open Dataset of User Preferences for Text-to-Image Generation

Researchers From Stanford And DeepMind Come Up With The Idea of Using Large Language Models LLMs as a Proxy Reward Function

Joscha Bach on Twitter