Our new X account is live! Follow @wizwand_team for updates

Financial Question Answering on FinanceIQ

73.6Accuracy (%)

Re-TASK (Full)

Updated 4d ago

Evaluation Results

Method	Links
Re-TASK (Full) 2024.08		73.6	-
Step-Back 2024.08		66.85	-
Plan-and-Solve 2024.08		66.29	-
Re-TASK (+K0) 2024.08		66.29	-
One-shot CoT 2024.08		64.33	-
Three-shot CoT 2024.08		63.82	-
Zero-shot CoT + SC 2024.08		61.8	-
Re-TASK (Lite) 2024.08		61.8	-
Zero-shot CoT 2024.08		53.93	-
Re-TASK (Full) 2024.08		52.81	-
Re-TASK (Full) 2024.08		51.69	-
Re-TASK (Lite) 2024.08		49.44	-
Zero-shot CoT + SC 2024.08		46.63	-
One-shot CoT 2024.08		46.07	-
Three-shot CoT 2024.08		46.07	-
Step-Back 2024.08		44.38	-
Zero-shot CoT 2024.08		43.82	-
Plan-and-Solve 2024.08		41.01	-
Re-TASK (+K0) 2024.08		39.89	-
Re-TASK (Lite) 2024.08		38.2	-
Zero-shot CoT 2024.08		36.52	-
One-shot CoT 2024.08		34.69	-
Zero-shot CoT + SC 2024.08		34.27	-
Three-shot CoT 2024.08		34.27	-
Re-TASK (+K0) 2024.08		34.27	-
Step-Back 2024.08		30.9	-
Plan-and-Solve 2024.08		30.34	-
Zero-shot CoT + SC 2024.08		-	2.81
One-shot CoT 2024.08		-	3.6
Three-shot CoT 2024.08		-	3.3
Plan-and-Solve 2024.08		-	1.12
Step-Back 2024.08		-	2.62
Re-TASK (+K0) 2024.08		-	2.06
Re-TASK (Lite) 2024.08		-	5.06
Re-TASK (Full) 2024.08		-	14.61