Skip to content
conftrace_
2026 EMNLP EMNLP 2026

MARD: A Multi-Agent Reinforced Debiasing Framework for Large Language Models

AbstractMitigating social bias in large language models (LLMs) has become an increasingly important research objective. However, existing methods often rely on prompt-level control or optimize bias-aware guidance and final prediction within a single model, which may lead to coupled optimization and task interference. To address these limitations, we propose MARD, a Multi-Agent Reinforced Debiasing framework that decomposes debiasing into two role-specialized agents. The Bias-Guard Agent identifies stereotype cues and generates bias-control guidance, while the Evidence-Decision Agent produces evidence-grounded predictions without relying on identity-based assumptions. MARD further introduces retrieved group-related references as negative warning signals to expose potential stereotype cues, and adopts an alternating GRPO optimization strategy to improve role specialization. Experiments on five datasets across four task types demonstrate that MARD achieves strong overall debiasing performance. Further analyses highlight the effectiveness of role decoupling, alternating optimization, format control, and retrieved warning references.

More from EMNLP 2026

All EMNLP 2026 papers →