Fairness-Aware Calibration of LLM Evaluators Using Tokenized Disclosures

Fine-tune LLM evaluators to avoid penalizing tokenized AI assistance disclosures and to remove demographic interaction effects documented in prior work. This approach operationalizes fairness by design and closes the loop between disclosure design and algorithmic judgment in ranking, hiring, and review systems.

References:

1. Penalizing Transparency? How AI Disclosure and Author Demographics Shape Human and AI Judgments About Writing. Inyoung Cheong, Alicia Guo, Mina Lee, Zhehui Liao, Kowe Kadoma, Dongyoung Go, Joseph Chee Chang, Peter Henderson, Mor Naaman, Amy X. Zhang (2025). arXiv.org.