# MABPD: Multi-Agent Bias Probing & Detection via Structured Argument Debate

arXiv: https://arxiv.org/abs/2609.04841 · HTML: https://mlsanity.com/paper/2609.04841/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use.
- Builds on 19 of its 42 references that are in the map.
- First posted on arXiv 4 Sep 2026.

## Placed in

- LLM agents and tool use / [Social simulation and multi-agent debate](https://mlsanity.com/agents/social-simulation-and-multi-agent-debate/)

## Builds on (most cited first, 19 of 19)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Improving Factuality and Reasoning in Language Models through Multiagent Debate](https://mlsanity.com/paper/2305.14325/) (2023)
- [Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate](https://mlsanity.com/paper/2305.19118/) (2023)
- [Why Do Multi-Agent LLM Systems Fail?](https://mlsanity.com/paper/2503.13657/) (2025)
- [Neural Media Bias Detection Using Distant Supervision With BABE -- Bias Annotations By Experts](https://mlsanity.com/paper/2209.14557/) (2022)
- [Investigating Bias in LLM-Based Bias Detection: Disparities between LLMs and Human Perception](https://mlsanity.com/paper/2403.14896/) (2024)
- [The Truth Becomes Clearer Through Debate! Multi-Agent Systems with Large Language Models Unmask Fake News](https://mlsanity.com/paper/2505.08532/) (2025)
- [Multi-Agent Debate for LLM Judges with Adaptive Stability Detection](https://mlsanity.com/paper/2510.12697/) (2025)
- [Enhancing Multi-Agent Debate System Performance via Confidence Expression](https://mlsanity.com/paper/2509.14034/) (2025)
- [Introducing MBIB -- the first Media Bias Identification Benchmark Task and Dataset Collection](https://mlsanity.com/paper/2304.13148/) (2023)
- [Neural Based Statement Classification for Biased Language](https://mlsanity.com/paper/1811.05740/) (2018)
- [AI Usage Cards: Responsibly Reporting AI-generated Content](https://mlsanity.com/paper/2303.03886/) (2023)
- [MALLM: Multi-Agent Large Language Models Framework](https://mlsanity.com/paper/2509.11656/) (2025)
- [Debate-to-Detect: Reformulating Misinformation Detection as a Real-World Debate with Large Language Models](https://mlsanity.com/paper/2505.18596/) (2025)
- [Towards A Reliable Ground-Truth For Biased Language Detection](https://mlsanity.com/paper/2112.07421/) (2021)
- [MAGPIE: Multi-Task Media-Bias Analysis Generalization for Pre-Trained Identification of Expressions](https://mlsanity.com/paper/2403.07910/) (2024)

## Abstract (arXiv)

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.