Hallucination-minimized Data-to-answer Framework for Financial Decision-makers

TL;DR

This paper introduces a novel framework that minimizes hallucinations in data-to-answer tasks for financial decision-making by classifying queries, retrieving relevant data chunks, and applying multi-metric scoring to ensure response confidence.

Contribution

The work presents a Langchain-based hierarchical data transformation and scoring system that significantly reduces hallucinations in LLM-generated answers in financial domains.

Findings

Achieves over 90% confidence scores in response accuracy.

Effectively classifies user queries by intention for targeted data retrieval.

Framework adaptable to other analytical domains like sales and payroll.

Abstract

Large Language Models (LLMs) have been applied to build several automation and personalized question-answering prototypes so far. However, scaling such prototypes to robust products with minimized hallucinations or fake responses still remains an open challenge, especially in niche data-table heavy domains such as financial decision making. In this work, we present a novel Langchain-based framework that transforms data tables into hierarchical textual data chunks to enable a wide variety of actionable question answering. First, the user-queries are classified by intention followed by automated retrieval of the most relevant data chunks to generate customized LLM prompts per query. Next, the custom prompts and their responses undergo multi-metric scoring to assess for hallucinations and response confidence. The proposed system is optimized with user-query intention classification, advanced prompting, data scaling capabilities and it achieves over 90% confidence scores for a variety of user-queries responses ranging from {What, Where, Why, How, predict, trend, anomalies, exceptions} that are crucial for financial decision making applications. The proposed data to answers framework can be extended to other analytical domains such as sales and payroll to ensure optimal hallucination control guardrails.