MirageBench

What is MirageBench?

Evaluates LLM personalization faithfulness: 150 personas, 6 tasks, four-way faithfulness taxonomy, 12 models across 7 families, 143,616 judged claims.

Released

2026-08-05

Evaluates

General AI, Language & Knowledge, cs.CL

Openness

unknown

Importer

Claire Radar

Review status

ai-reviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2608.04570

MirageBench paper

• Paper https://arxiv.org/abs/2608.04570

Which sources cite MirageBench?

• MirageBench

Related General AI, Language & Knowledge, cs.CL benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 3DCodeBench
• 5-Dialects-BN
• A Large-Scale Dataset and Benchmark
• A3S-Bench
• AARRI-Bench
• ABC-Bench
• ActBench