From Fragments to Facts: A Curriculum-Driven DPO Approach for Generating Hindi News Veracity Explanations
This addresses misinformation detection for low-resource languages like Hindi, though it is an incremental improvement by adapting existing methods to a new domain.
The paper tackles the problem of generating reliable news veracity explanations for Hindi, an under-represented language, by proposing a curriculum-driven DPO framework that integrates Actuality and Finesse parameters into the loss function. Experiments with models like Mistral and mBART show the approach effectively generates coherent and contextually relevant explanations.
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.