arXiv cs.CL· Noah Dasanaike·· 2 天前AI 评分30
大语言模型可从个人记录中朴素推断族裔
Large Language Models Naively Recover Ethnicity from Individual Records
AI 导读
一项已撤回的研究显示，大语言模型无需额外训练数据即可从姓名推断族裔，在佛罗里达和北卡罗来纳选民样本上准确率达 84.7%，超过 BISG 的 68.2%。测试涵盖 Gemini 3 Flash、GPT-4o、DeepSeek v3.2 和 GLM-4.7 等六个模型，启用扩展推理可提升 1-3 个百分点，加入党派登记等元数据可达 86.7%。
正文 · 原文
This paper has been withdrawn by Noah Dasanaike
No PDF available, click to view other formats
Abstract:I demonstrate that large language models can infer ethnicity from names with accuracy exceeding that of Bayesian Improved Surname Geocoding (BISG) without additional training data, enabling inference outside the United States and to contextually appropriate classification categories. Using stratified samples from Florida and North Carolina voter files with self-reported race, LLM-based classification achieves up to 84.7% accuracy, outperforming BISG (68.2%) on balanced samples. I test six models including Gemini 3 Flash, GPT-4o, and open-source alternatives such as DeepSeek v3.2 and GLM-4.7. Enabling extended reasoning can improve accuracy by 1-3 percentage points, though effects vary across contexts; including metadata such as party registration reaches 86.7%. LLM classification also reduces the income bias inherent in BISG, where minorities in wealthier neighborhoods are systematically misclassified as White. I further validate using Lebanese voter registration with religious sect (64.3% accuracy), Indian MPs from reserved constituencies (99.2%), and Indian land records with caste classification (74.0%). Aggregate validation across India, Uganda, Nepal, Armenia, Chile, and Costa Rica using original full-count voter rolls demonstrates that the method recovers known population distributions where naming conventions are distinctive. For large-scale applications, small transformer models fine-tuned on LLM labels exceed BISG accuracy while enabling local deployment at no cost.
| Comments: | This paper has been superseded by arXiv:2610.19048 (merged from arXiv:2601.21132 and arXiv:2601.21138) |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2601.21132 [cs.CL] |
| (or arXiv:2601.21132v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2601.21132 arXiv-issued DOI via DataCite |
Submission history
From: Noah Dasanaike [view email]
[v1]
Thu, 29 Jan 2026 00:17:58 UTC (33 KB)
[v2]
Thu, 15 Oct 2026 01:58:16 UTC (1 KB) (withdrawn)
来源：arXiv cs.CL · arxiv.org