Study Finds MLLM Judges Show Cultural Bias Under Ambiguity
Scoring bias in judge models flows directly into training signal and leaderboard rankings, so skewed calibration quietly shapes which outputs get rewarded. Teams using MLLM-as-a-judge pipelines should audit scale usage and split alignment reports by reference culture rather than averaging. Anyone building evaluation or RLHF infrastructure for products serving both Western and Chinese audiences has a concrete methodology to copy here.
A paper titled "Jury Duty: Calibration and Orientation Failures in MLLM-as-a-Judge Under Cultural Ambiguity" tests how multimodal LLM judges score culturally contested content. The researchers built VOIR DIRE, a benchmark of 626 paired U.S. and mainland Chinese image items spanning food, fashion, and architecture. They evaluated six MLLM judges and found the models rarely use the low end of the rating scale, tilting contested items toward one cultural reading. Persona prompts only partly correct the tilt, and in-context examples deepen it. The paper recommends reporting alignment against each cultural reference pool separately.
⚖️ Jury Duty: Calibration and Orientation Failures in MLLM-as-a-Judge Under Cultural Ambiguity: https://t.co/iYleB4mlas
AI judges score the outputs that become training signal and leaderboard rankings. We tested whose judgment they reflect when cultures disagree.
→ VOIR DIRE: 626 paired U.S. and mainland Chinese image items across food, fashion, and architecture → Six MLLM judges rarely use the low end of the scale, which tilts contested items toward one cultural reading → Persona prompts only partly correct the tilt, and in-context examples deepen it
Our recommendation: report alignment against each reference pool separately.
#FutureOfAI #EnterpriseAI #EMNLP2026
