DFAH-Bench

What is DFAH-Bench?

DFAH-Bench measures whether financial agents reproduce decisions and observable execution paths across qualified replay groups.

Released

2026-06-10

Evaluates

Agents, Finance, Language & Knowledge

Openness

unknown

Importer

Claire Radar

Review status

ai-reviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2607.20491

DFAH-Bench paper and code

• Paper https://arxiv.org/abs/2607.20491
• Code https://github.com/ibm-client-engineering/output-drift-financial-llms

Which sources cite DFAH-Bench?

• DFAH-Bench

Related Agents, Finance, Language & Knowledge benchmarks

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