Rubrics derived from authentic financial deliverables beat prompt-only rubrics on role-specialized tasks (99.1% vs 78.0% coverage) while matching them on conventional tasks.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
FinProBench: Evaluating Financial AI Agents with Role-Grounded Rubrics Derived from Professional Deliverables
Rubrics derived from authentic financial deliverables beat prompt-only rubrics on role-specialized tasks (99.1% vs 78.0% coverage) while matching them on conventional tasks.