Get Started
Home
Topics
Search
Library
Research questionHow can subtle media bias be detected without task-specific annotated training data?Media bias can appear through loaded language, selective framing, and strategic omission, making it difficult to identify reliably. These cues have traditionally been detected using large collections of annotated examples.
AI
Evaluation & Benchmarks
Machine Learning
Multi-agent Systems
Natural Language Processing
Reasoning
Research Paper
Latest papersRecent research connected to this question, newest first.MABPD: Multi-Agent Bias Probing & Detection via Structured Argument DebateThe source evaluates a structured multi-agent LLM pipeline for sentence- and article-level media-bias detection without task-specific training or threshold tuning. Evidence comes from the BABE benchmark and zero-shot evaluation on the SemEval 2019 HyperPartisan corpus; broader robustness and deployment constraints are not established.research paper · Sep 4, 2026
Related questions
How can we reduce fine-tuning-induced shortcut reliance on underrepresented groups without retraining or group labels?How can bias evaluations of large language models diagnose affected groups and reasons behind biased outputs?How can detect-to-track pipelines adapt to new visual domains without target-domain labels?How can spoken language detection adapt to underrepresented accents under low-resource constraints without overfitting?