What Current AI Benchmarks Leave Unmeasured: Modality, Search, Citations, and Implications (for Safety Evaluations)
Published in AAAI/ACM Conference on AI, Ethics, and Society (AIES), 2026
Code and data: github.com/seerocode/aies-llm-modality-audit
Recommended citation: Ro Encarnación, Tina Behzad, Emma Lurie, and Danaé Metaxa. (2026). What Current AI Benchmarks Leave Unmeasured: Modality, Search, Citations, and Implications (for Safety Evaluations). AAAI/ACM Conference on AI, Ethics, and Society (AIES).
Download Paper