ACL Findings2025
Revealing Hidden Mechanisms of Cross-Country Content Moderation with Natural Language Processing
Neemesh Yadav, Jiarui Liu, Francesco Ortu, Roya Ensafi, Zhijing Jin, Rada Mihalcea
Abstract
The ability of Natural Language Processing (NLP) methods to categorize text into multiple classes has motivated their use in online content moderation tasks, such as hate speech and fake news detection. However, there is limited understanding of how or why these methods make such decisions, or why certain content is moderated in the first place. To investigate the hidden mechanisms behind content moderation, we explore multiple directions: 1) training classifiers to reverse-engineer content moderation decisions across countries; 2) explaining content moderation decisions by analyzing Shapley values and LLM-guided explanations. Our primary focus is on content moderation decisions made across countries, using pre-existing corpora sampled from the Twitter Stream Grab. Our experiments reveal interesting patterns in censored posts, both across countries and over time. Through human evaluations of LLMgenerated explanations across three LLMs, we assess the effectiveness of using LLMs in content moderation. Finally, we discuss potential future directions, as well as the limitations and ethical considerations of this work. 1
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- Towards Automated Circuit Discovery for Mechanistic Interpretability cites Locating and Editing Factual Associations in GPT
- Locating and Editing Factual Associations in GPT cites Investigating Gender Bias in Language Models Using Causal Mediation Analysis
Builds on10
- Locating and Editing Factual Associations in GPTKevin Meng, David Bau, Alex Andonian, Yonatan BelinkovNeurIPS 2022 · 3,415 citations
- Pythia: A Suite for Analyzing Large Language Models Across Training and ScalingStella Biderman, Hailey Schoelkopf, Quentin Gregory Anthony, Herbie Bradley et al.ICML 2023 · 1,822 citations
- Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought PromptingMiles Turpin, Julian Michael, Ethan Perez, Samuel R. BowmanNeurIPS 2023 · 1,792 citations
- Towards Automated Circuit Discovery for Mechanistic InterpretabilityArthur Conmy, Augustine N. Mavor-Parker, Aengus Lynch, Stefan Heimersheim et al.NeurIPS 2023 · 861 citations
- Investigating Gender Bias in Language Models Using Causal Mediation AnalysisJesse Vig, Sebastian Gehrmann, Yonatan Belinkov, Sharon Qian et al.NeurIPS 2020 · 851 citations
Related papers
- Evaluating Large Language Models for Detecting AntisemitismJay Patel, Hrudayangam Mehta, Jeremy BlackburnEMNLP 2025 · 1 citation
- MoMoE: Mixture of Moderation Experts Framework for AI-Assisted Online GovernanceAgam Goyal, Xianyang Zhan, Yilun Chen, Koustuv Saha et al.EMNLP 2025
- Using Pre-Trained Language Models for Producing Counter Narratives Against Hate Speech: a Comparative StudySerra Sinem Tekiroglu, Helena Bonaldi, Margherita Fanton, Marco GueriniACL Findings 2022 · 16 citations
- There is No War in Ba Sing Se: A Global Analysis of Content Moderation in Large Language ModelsFriedemann Lipphardt, Moonis Ali, Martin Banzer, Anja Feldmann et al.NDSS 2026 · 1 citation
- SMARTER: A Data-efficient Framework to Improve Toxicity Detection with Explanation via Self-augmenting Large Language ModelsHuy Nghiem, Advik Sachdeva, Hal Daumé IIIACL 2026 · 1 citation
