From Fragments to Facts: A Curriculum-Driven DPO Approach for Generating Hindi News Veracity Explanations

TL;DR

This paper presents a curriculum-driven DPO framework for generating reliable Hindi news explanations, addressing misinformation in low-resource languages by aligning machine outputs with human reasoning.

Contribution

It introduces a novel DPO-based approach with curriculum learning and new parameters, improving explanation quality for Hindi news verification.

Findings

Effective in generating coherent explanations with LLMs and PLMs

Enhances explanation quality using Actuality and Finesse parameters

Scalable method for low-resource language misinformation detection

Abstract

In an era of rampant misinformation, generating reliable news explanations is vital, especially for under-represented languages like Hindi. Lacking robust automated tools, Hindi faces challenges in scaling misinformation detection. To bridge this gap, we propose a novel framework integrating Direct Preference Optimization (DPO) with curriculum learning to align machine-generated explanations with human reasoning. Fact-checked explanations from credible sources serve as preferred responses, while LLM outputs highlight system limitations and serve as non-preferred responses. To refine task-specific alignment, we introduce two key parameters -- Actuality and Finesse -- into the DPO loss function, enhancing explanation quality and consistency. Experiments with LLMs (Mistral, Llama, Gemma) and PLMs (mBART, mT5) confirm the framework's effectiveness in generating coherent, contextually relevant explanations. This scalable approach combats misinformation and extends automated explanation generation to low-resource languages.