YinYang-Align: A new Benchmark for Competing Objectives and Introducing Multi-Objective Preference based Text-to-Image Alignment

Precise alignment in Text-to-Image (T2I) systems is crucial for generating visuals that accurately reflect user intent while adhering to ethical and policy standards. Recent controversies, such as backlash against Google Geminigenerated images, underscore the need for stronger alignment mechanisms. Building on alignment successes in Large Language Models (LLMs), this paper introduces YinYan-gAlign, a benchmarking framework designed to evaluate and optimize T2I systems across six inherently contradictory objectives. These objectives highlight fundamental trade-offs, such as balancing faithfulness to prompts with artistic freedom and maintaining cultural sensitivity without compromising creativity.