AuthorityBench

What is AuthorityBench?

AuthorityBench evaluates how citation-based authority signals affect epistemic behavior in LLMs using 220,564 prompts across four domains, with a 2x2 factorial…

Released

2026-06-11

Evaluates

General AI, Safety & Trustworthiness, cs.LG

Openness

unknown

Importer

Claire Radar

Review status

unreviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2606.13104

AuthorityBench paper and code

• Paper https://arxiv.org/abs/2606.13104
• Code https://github.com/floating-reeds/AuthorityBench

Which sources cite AuthorityBench?

• AuthorityBench

Related General AI, Safety & Trustworthiness, cs.LG 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
• ActBench
• ActiveSciBench