Artificial Inteligence feed for Tuesday, December 10, 2024

• Study: Some language reward models exhibit political bias
• Enabling AI to explain its predictions in plain language