🎉 Examining Alignment of Large Language Models through Representative Heuristics: The Case of Political Stereotypes is accepted ICLR 2025