{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/persistent-anti-muslim-bias-in-large-language","title":"Persistent Anti-Muslim Bias in Large Language Models","arxiv_id":"2101.05783","date":"2021-01-14","proceeding":null,"authors":["Abubakar Abid","Maheen Farooqi","James Zou"],"abstract":"It has been observed that large-scale language models capture undesirable societal biases, e.g. relating to race and gender; yet religious bias has been relatively unexplored. We demonstrate that GPT-3, a state-of-the-art contextual language model, captures persistent Muslim-violence bias. We probe GPT-3 in various ways, including prompt completion, analogical reasoning, and story generation, to understand this anti-Muslim bias, demonstrating that it appears consistently and creatively in different uses of the model and that it is severe even compared to biases about other religious groups. For instance, \"Muslim\" is analogized to \"terrorist\" in 23% of test cases, while \"Jewish\" is mapped to \"money\" in 5% of test cases. We quantify the positive distraction needed to overcome this bias with adversarial text prompts, and find that use of the most positive 6 adjectives reduces violent completions for \"Muslims\" from 66% to 20%, but which is still higher than for other religious groups.","url_abs":"https://arxiv.org/abs/2101.05783v2","url_pdf":"https://arxiv.org/pdf/2101.05783v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"persistent-anti-muslim-bias-in-large-language","repo_url":"https://github.com/clip-italian/clip-italian","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"jax","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"adversarial-text","task_name":"Adversarial Text"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"story-generation","task_name":"Story Generation"}],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"cosine-annealing","method_name":"Cosine Annealing"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"gpt-3","method_name":"GPT-3"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"linear-warmup-with-cosine-annealing","method_name":"Linear Warmup With Cosine Annealing"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"weight-decay","method_name":"Weight Decay"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2101.05783","atlas_url":"https://app.syntology.ai/?focus=2101.05783","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}