Media bias in news articles operates through subtle linguistic cues—loaded language, selective framing, and strategic omission—that resist single-model detection and have traditionally required large annotated corpora for supervised training. We ask whether structured multi-agent deliberation can serve as a principled, training-free alternative to supervised classification for this task. We introduce MABPD (Multi-Agent Bias Probing Detection), a pipeline in which three specialized LLM agents analyze an article from complementary perspectives and resolve disagreements through a Structured Argument Debate (SAD) protocol. SAD implements a domain-motivated asymmetric burden of proof—biased claims without grounded textual evidence carry zero weight—combined with role-weighted voting and post-consensus verification, replacing task-specific supervised decision boundaries with explicit deliberative structure. Ablation confirms that this structured deliberation, not mere agent parallelism, drives performance: removing the debate module reduces F1 by up to 10.6 points. On the BABE benchmark (4,121 expert-annotated sentences), MABPD achieves 83.4% macro F1 on the held-out test split—within 0.7 percentage points (pp) of the supervised SOTA (MAGPIE, 84.1% macro F1; Horych et al., 2024)—without any task-specific training or threshold tuning on annotated data. Cross-dataset evaluation on the SemEval 2019 HyperPartisan corpus (644 articles) yields 75.0% zero-shot accuracy, within 7.2 pp of the supervised SOTA accuracy (82.2%; Kiesel et al. 2019), confirming transfer across annotation regimes. We release the full pipeline and evaluation code.

The recent advances in AI made generation of questions from natural language text possible, the approach is completely excludes human in the loop while generating the appropriate questions which improves the students learning engagement. The ever growing amount of educational content renders it increasingly difficult to manually generate sufficient practice or quiz questions to accompany it. Reading comprehension can be improved by asking the right questions, So, this work offers a Transformer based question generation model for autonomously producing quiz questions from educational information, such as eBooks. This work proposes an contrastive training approach for ‘Text-to-Text Transfer Transformer’ (T5) model where the model (T5-eQG) creates the summarized text for the input document and then automatically generates the questions. Our model shows promising results over earlier neural network-based and rules-based models for question generating task on benchmark datasets and NCERT eBooks.