% SocioLens, LD-07 version v0.2: https://sociolens-web.pages.dev/debates/ld-07/v0.2/
% Generated from SocioLens source labels. Authors and years are not stored as separate fields,
% so they are not filled in here, and a title is given only where the label states one;
% the full label is in each note. For complete records, import the identifiers file into
% Zotero (Add Item by Identifier). "SocioLens synthesis" entries are the project's own
% synthesis documents, not primary evidence or scholarship.

@misc{src_ld07_egami_downstream_inference_2023,
  title = {Using Imperfect Surrogates for Downstream Inference: Design-based Supervised Learning for Social Science Applications of Large Language Models},
  doi = {10.52202/075280-3000},
  url = {https://doi.org/10.52202/075280-3000},
  howpublished = {Source type: peer reviewed conference paper},
  note = {Egami, Hinck, Stewart \& Wei, “Using Imperfect Surrogates for Downstream Inference: Design-based Supervised Learning for Social Science Applications of Large Language Models,” NeurIPS 2023},
  annote = {SocioLens source SRC-LD07-EGAMI-DOWNSTREAM-INFERENCE-2023: https://sociolens-web.pages.dev/sources/src-ld07-egami-downstream-inference-2023/v1.0/}
}

@misc{src_ld07_baumann_llm_hacking_2025,
  title = {Large Language Model Hacking},
  eprint = {2509.08825},
  archiveprefix = {arXiv},
  url = {https://arxiv.org/abs/2509.08825v2},
  howpublished = {Source type: preprint},
  note = {Baumann, Röttger, Urman, Wendsjö, Plaza-del-Arco, Gruber \& Hovy, “Large Language Model Hacking,” arXiv:2509.08825v2},
  annote = {SocioLens source SRC-LD07-BAUMANN-LLM-HACKING-2025: https://sociolens-web.pages.dev/sources/src-ld07-baumann-llm-hacking-2025/v1.0/}
}

@misc{src_ld07_calderon_alternative_annotator_2025,
  title = {The Alternative Annotator Test for LLM-as-a-Judge},
  doi = {10.18653/v1/2025.acl-long.782},
  url = {https://doi.org/10.18653/v1/2025.acl-long.782},
  howpublished = {Source type: peer reviewed conference paper},
  note = {Calderon, Reichart \& Dror, “The Alternative Annotator Test for LLM-as-a-Judge,” ACL 2025},
  annote = {SocioLens source SRC-LD07-CALDERON-ALTERNATIVE-ANNOTATOR-2025: https://sociolens-web.pages.dev/sources/src-ld07-calderon-alternative-annotator-2025/v1.0/}
}

@misc{src_ld07_nakamura_observational_equivalence_2026,
  title = {Observational Equivalence of LLM and Human Annotation},
  eprint = {2609.22133},
  archiveprefix = {arXiv},
  url = {https://arxiv.org/abs/2609.22133},
  howpublished = {Source type: preprint},
  note = {Nakamura, Tan \& Yean, “Observational Equivalence of LLM and Human Annotation,” arXiv:2609.22133},
  annote = {SocioLens source SRC-LD07-NAKAMURA-OBSERVATIONAL-EQUIVALENCE-2026: https://sociolens-web.pages.dev/sources/src-ld07-nakamura-observational-equivalence-2026/v1.0/}
}

@misc{src_ld07_ashokkumar_nature_2026,
  title = {Large language models can predict the results of social science experiments},
  doi = {10.1038/s41586-026-10742-x},
  url = {https://doi.org/10.1038/s41586-026-10742-x},
  howpublished = {Source type: peer reviewed journal article},
  note = {Ashokkumar, Hewitt, Ghezae \& Willer, “Large language models can predict the results of social science experiments,” Nature 656:115–122 (2026)},
  annote = {SocioLens source SRC-LD07-ASHOKKUMAR-NATURE-2026: https://sociolens-web.pages.dev/sources/src-ld07-ashokkumar-nature-2026/v1.0/}
}

@misc{src_ld07_bisbee_synthetic_replacements_2024,
  title = {Synthetic Replacements for Human Survey Data? The Perils of Large Language Models},
  doi = {10.1017/pan.2024.5},
  url = {https://doi.org/10.1017/pan.2024.5},
  howpublished = {Source type: peer reviewed journal article},
  note = {Bisbee, Clinton, Dorff, Kenkel \& Larson, “Synthetic Replacements for Human Survey Data? The Perils of Large Language Models,” Political Analysis 32(4):401–416 (2024)},
  annote = {SocioLens source SRC-LD07-BISBEE-SYNTHETIC-REPLACEMENTS-2024: https://sociolens-web.pages.dev/sources/src-ld07-bisbee-synthetic-replacements-2024/v1.0/}
}

@misc{src_ld07_xie_ssdatabench_2026,
  title = {Evaluating the statistical realism of LLM-generated social science data},
  doi = {10.1073/pnas.2538145123},
  url = {https://doi.org/10.1073/pnas.2538145123},
  howpublished = {Source type: peer reviewed journal article},
  note = {Xie et al., “Evaluating the statistical realism of LLM-generated social science data,” PNAS 123(19):e2538145123 (2026)},
  annote = {SocioLens source SRC-LD07-XIE-SSDATABENCH-2026: https://sociolens-web.pages.dev/sources/src-ld07-xie-ssdatabench-2026/v1.0/}
}

@misc{src_r03_wang_morgenstern_dickerson_nmi_2025,
  howpublished = {Source type: peer reviewed journal article},
  note = {Wang et al. NMI 2025},
  annote = {SocioLens source SRC-R03-WANG-MORGENSTERN-DICKERSON-NMI-2025: https://sociolens-web.pages.dev/sources/src-r03-wang-morgenstern-dickerson-nmi-2025/v1.0/}
}

@misc{src_ld07_gong_huang_cgss_2025,
  title = {整体事实还是有偏样本——基于大语言模型生成数据的测量},
  url = {https://www.xsyk021.com/article/id/4c1886eb-c385-4019-92ff-d1cd75547fa5},
  howpublished = {Source type: peer reviewed journal article},
  note = {Gong Weigang \& Huang Siyuan, “整体事实还是有偏样本——基于大语言模型生成数据的测量,” Academic Monthly 57(6):123–136 (2025)},
  annote = {SocioLens source SRC-LD07-GONG-HUANG-CGSS-2025: https://sociolens-web.pages.dev/sources/src-ld07-gong-huang-cgss-2025/v1.0/}
}

@misc{src_ld07_ling_cui_silicon_samples_2025,
  title = {复现的限度——‘硅基样本’的可能与可为},
  howpublished = {Source type: peer reviewed journal article},
  note = {Ling Wanying \& Cui Sizhan, “复现的限度——‘硅基样本’的可能与可为,” Journal of Intelligent Society 4(2):179–201 (2025)},
  annote = {SocioLens source SRC-LD07-LING-CUI-SILICON-SAMPLES-2025: https://sociolens-web.pages.dev/sources/src-ld07-ling-cui-silicon-samples-2025/v1.0/}
}

@misc{src_ld07_li_ji_realism_ate_2026,
  title = {Statistical realism is not evidence that LLMs can estimate treatment effects in social science experiments},
  eprint = {2604.02458},
  archiveprefix = {arXiv},
  url = {https://arxiv.org/abs/2604.02458},
  howpublished = {Source type: preprint},
  note = {Li \& Ji, “Statistical realism is not evidence that LLMs can estimate treatment effects in social science experiments,” arXiv:2604.02458},
  annote = {SocioLens source SRC-LD07-LI-JI-REALISM-ATE-2026: https://sociolens-web.pages.dev/sources/src-ld07-li-ji-realism-ate-2026/v1.0/}
}

@misc{src_ld07_westwood_survey_threat_2025,
  title = {The potential existential threat of large language models to online survey research},
  doi = {10.1073/pnas.2518075122},
  url = {https://doi.org/10.1073/pnas.2518075122},
  howpublished = {Source type: peer reviewed journal article},
  note = {Westwood, “The potential existential threat of large language models to online survey research,” PNAS 122(47):e2518075122 (2025)},
  annote = {SocioLens source SRC-LD07-WESTWOOD-SURVEY-THREAT-2025: https://sociolens-web.pages.dev/sources/src-ld07-westwood-survey-threat-2025/v1.0/}
}

@misc{src_ld07_westwood_frederick_reply_2026,
  title = {Reply to Van der Stigchel et al.: Empirical evidence that AI survey contamination is real and substantial},
  doi = {10.1073/pnas.2537420123},
  url = {https://doi.org/10.1073/pnas.2537420123},
  howpublished = {Source type: peer reviewed journal article},
  note = {Westwood \& Frederick, “Reply to Van der Stigchel et al.: Empirical evidence that AI survey contamination is real and substantial,” PNAS 123(8):e2537420123 (2026)},
  annote = {SocioLens source SRC-LD07-WESTWOOD-FREDERICK-REPLY-2026: https://sociolens-web.pages.dev/sources/src-ld07-westwood-frederick-reply-2026/v1.0/}
}

@misc{src_ld07_asher_keystroke_2026,
  title = {Chatbots Are Undermining Crowdsourced Research in the Behavioral Sciences},
  doi = {10.1177/25152459261424723},
  url = {https://doi.org/10.1177/25152459261424723},
  howpublished = {Source type: peer reviewed journal article},
  note = {Asher, Gold, Chen \& Carvalho, “Chatbots Are Undermining Crowdsourced Research in the Behavioral Sciences,” AMPPS 9(1) (2026)},
  annote = {SocioLens source SRC-LD07-ASHER-KEYSTROKE-2026: https://sociolens-web.pages.dev/sources/src-ld07-asher-keystroke-2026/v1.0/}
}

@misc{src_ld07_zhang_homogenization_2025,
  title = {Generative AI Meets Open-Ended Survey Responses},
  doi = {10.1177/00491241251327130},
  url = {https://doi.org/10.1177/00491241251327130},
  howpublished = {Source type: peer reviewed journal article},
  note = {Zhang, Xu \& Alvero, “Generative AI Meets Open-Ended Survey Responses,” Sociological Methods \& Research 54(3):1197–1242 (2025)},
  annote = {SocioLens source SRC-LD07-ZHANG-HOMOGENIZATION-2025: https://sociolens-web.pages.dev/sources/src-ld07-zhang-homogenization-2025/v1.0/}
}

@misc{src_ld07_veselovsky_crowdwork_2025,
  title = {Prevalence and prevention of large language model use in crowd work},
  doi = {10.1145/3685527},
  url = {https://doi.org/10.1145/3685527},
  howpublished = {Source type: peer reviewed journal article},
  note = {Veselovsky, Horta Ribeiro, Cozzolino, Gordon, Rothschild \& West, “Prevalence and prevention of large language model use in crowd work,” CACM 2025},
  annote = {SocioLens source SRC-LD07-VESELOVSKY-CROWDWORK-2025: https://sociolens-web.pages.dev/sources/src-ld07-veselovsky-crowdwork-2025/v1.0/}
}

@misc{src_ld07_chen_ai_agent_responding_2026,
  title = {Estimating the threat of AI-agent responding across online survey platforms},
  doi = {10.31234/osf.io/xcg26_v1},
  url = {https://doi.org/10.31234/osf.io/xcg26_v1},
  howpublished = {Source type: preprint},
  note = {Chen, Urminsky et al., “Estimating the threat of AI-agent responding across online survey platforms,” PsyArXiv 2026},
  annote = {SocioLens source SRC-LD07-CHEN-AI-AGENT-RESPONDING-2026: https://sociolens-web.pages.dev/sources/src-ld07-chen-ai-agent-responding-2026/v1.0/}
}

@misc{src_ld07_gordon_prevalence_2026,
  title = {AI Agent Prevalence and Data Quality Across Multiple Online Sample Providers},
  doi = {10.31234/osf.io/pvdjr_v2},
  url = {https://doi.org/10.31234/osf.io/pvdjr_v2},
  howpublished = {Source type: preprint},
  note = {Gordon et al., “AI Agent Prevalence and Data Quality Across Multiple Online Sample Providers,” PsyArXiv 2026},
  annote = {SocioLens source SRC-LD07-GORDON-PREVALENCE-2026: https://sociolens-web.pages.dev/sources/src-ld07-gordon-prevalence-2026/v1.0/}
}

@misc{src_ld07_xu_malkin_prompts_2026,
  title = {A Penny for Your Prompts: Experiments Detecting and Mitigating LLM Usage by Survey Respondents},
  eprint = {2607.00403},
  archiveprefix = {arXiv},
  url = {https://arxiv.org/abs/2607.00403},
  howpublished = {Source type: preprint},
  note = {Xu \& Malkin, “A Penny for Your Prompts: Experiments Detecting and Mitigating LLM Usage by Survey Respondents,” arXiv:2607.00403},
  annote = {SocioLens source SRC-LD07-XU-MALKIN-PROMPTS-2026: https://sociolens-web.pages.dev/sources/src-ld07-xu-malkin-prompts-2026/v1.0/}
}

@misc{src_ld07_lin_taxonomy_2026,
  title = {Artificial-Intelligence-Mediated Contamination in Online Research: Taxonomy, Risk Gradient, and Recommendations},
  doi = {10.1177/25152459261454231},
  url = {https://doi.org/10.1177/25152459261454231},
  howpublished = {Source type: peer reviewed journal article},
  note = {Lin, “Artificial-Intelligence-Mediated Contamination in Online Research: Taxonomy, Risk Gradient, and Recommendations,” AMPPS 9(3) (2026)},
  annote = {SocioLens source SRC-LD07-LIN-TAXONOMY-2026: https://sociolens-web.pages.dev/sources/src-ld07-lin-taxonomy-2026/v1.0/}
}

@misc{src_ld07_pan_xu_censorship_2026,
  title = {Political censorship in large language models originating from China},
  doi = {10.1093/pnasnexus/pgag013},
  url = {https://doi.org/10.1093/pnasnexus/pgag013},
  howpublished = {Source type: peer reviewed journal article},
  note = {Pan \& Xu, “Political censorship in large language models originating from China,” PNAS Nexus 5(2):pgag013 (2026)},
  annote = {SocioLens source SRC-LD07-PAN-XU-CENSORSHIP-2026: https://sociolens-web.pages.dev/sources/src-ld07-pan-xu-censorship-2026/v1.0/}
}

@misc{src_ld07_chen_zaharia_zou_drift_2024,
  title = {How is ChatGPT’s behavior changing over time?},
  doi = {10.1162/99608f92.5317da47},
  url = {https://doi.org/10.1162/99608f92.5317da47},
  howpublished = {Source type: peer reviewed journal article},
  note = {Chen, Zaharia \& Zou, “How is ChatGPT’s behavior changing over time?,” Harvard Data Science Review 6(2) (2024)},
  annote = {SocioLens source SRC-LD07-CHEN-ZAHARIA-ZOU-DRIFT-2024: https://sociolens-web.pages.dev/sources/src-ld07-chen-zaharia-zou-drift-2024/v1.0/}
}

@misc{src_ld07_shi_judging_the_judges_2025,
  title = {Judging the Judges: A Systematic Study of Position Bias in LLM-as-a-Judge},
  doi = {10.18653/v1/2025.ijcnlp-long.18},
  url = {https://doi.org/10.18653/v1/2025.ijcnlp-long.18},
  howpublished = {Source type: peer reviewed conference paper},
  note = {Shi et al., “Judging the Judges: A Systematic Study of Position Bias in LLM-as-a-Judge,” IJCNLP-AACL 2025},
  annote = {SocioLens source SRC-LD07-SHI-JUDGING-THE-JUDGES-2025: https://sociolens-web.pages.dev/sources/src-ld07-shi-judging-the-judges-2025/v1.0/}
}

@misc{src_ld07_panickssery_self_preference_2024,
  title = {LLM Evaluators Recognize and Favor Their Own Generations},
  url = {https://proceedings.neurips.cc/paper_files/paper/2024/hash/7f1f0218e45f5414c79c0679633e47bc-Abstract-Conference.html},
  howpublished = {Source type: peer reviewed conference paper},
  note = {Panickssery, Bowman \& Feng, “LLM Evaluators Recognize and Favor Their Own Generations,” NeurIPS 2024},
  annote = {SocioLens source SRC-LD07-PANICKSSERY-SELF-PREFERENCE-2024: https://sociolens-web.pages.dev/sources/src-ld07-panickssery-self-preference-2024/v1.0/}
}

@misc{src_ld07_than_future_of_coding_2025,
  title = {Updating ‘The Future of Coding’: Qualitative Coding with Generative Large Language Models},
  doi = {10.1177/00491241251339188},
  url = {https://doi.org/10.1177/00491241251339188},
  howpublished = {Source type: peer reviewed journal article},
  note = {Than, Fan, Law, Nelson \& McCall, “Updating ‘The Future of Coding’: Qualitative Coding with Generative Large Language Models,” Sociological Methods \& Research 54(3):849–888 (2025)},
  annote = {SocioLens source SRC-LD07-THAN-FUTURE-OF-CODING-2025: https://sociolens-web.pages.dev/sources/src-ld07-than-future-of-coding-2025/v1.0/}
}

@misc{src_ld07_plos_thematic_analysis_2026,
  doi = {10.1371/journal.pdig.0001189},
  url = {https://doi.org/10.1371/journal.pdig.0001189},
  howpublished = {Source type: peer reviewed journal article},
  note = {“Large language models for thematic analysis in healthcare research: A blinded mixed-methods comparison with human analysts,” PLOS Digital Health 2026},
  annote = {SocioLens source SRC-LD07-PLOS-THEMATIC-ANALYSIS-2026: https://sociolens-web.pages.dev/sources/src-ld07-plos-thematic-analysis-2026/v1.0/}
}

@misc{src_ld07_liu_agreement_is_not_quality_2026,
  title = {Agreement Is Not Quality: Blind Expert Verification of Human and LLM Qualitative Coding When Human Consensus Is Not Ground Truth},
  eprint = {2607.28890},
  archiveprefix = {arXiv},
  url = {https://arxiv.org/abs/2607.28890},
  howpublished = {Source type: preprint},
  note = {Liu et al., “Agreement Is Not Quality: Blind Expert Verification of Human and LLM Qualitative Coding When Human Consensus Is Not Ground Truth,” arXiv:2607.28890},
  annote = {SocioLens source SRC-LD07-LIU-AGREEMENT-IS-NOT-QUALITY-2026: https://sociolens-web.pages.dev/sources/src-ld07-liu-agreement-is-not-quality-2026/v1.0/}
}

@misc{src_ld07_yang_data_annotation_evaluation,
  title = {Data Annotation with Large Language Models: Lessons from a Large Empirical Evaluation},
  url = {https://www.eddieyang.net/research/llm_annotation.pdf},
  howpublished = {Source type: working paper},
  note = {Yang, Wang, Zhou \& Xu, “Data Annotation with Large Language Models: Lessons from a Large Empirical Evaluation,” conditionally accepted at AJPS; pre-publication manuscript},
  annote = {SocioLens source SRC-LD07-YANG-DATA-ANNOTATION-EVALUATION: https://sociolens-web.pages.dev/sources/src-ld07-yang-data-annotation-evaluation/v1.0/}
}

@misc{src_sl_argyle_busby_fulda_gubler_rytting_wingate_arxiv_2209_06899,
  title = {Out of One, Many: Using Language Models to Simulate Human Samples},
  eprint = {2209.06899},
  archiveprefix = {arXiv},
  url = {https://arxiv.org/abs/2209.06899v1},
  howpublished = {Source type: preprint},
  note = {Argyle, Busby, Fulda, Gubler, Rytting \& Wingate, “Out of One, Many: Using Language Models to Simulate Human Samples,” arXiv:2209.06899 v1, September 2022},
  annote = {SocioLens source SRC-SL-ARGYLE-BUSBY-FULDA-GUBLER-RYTTING-WINGATE-ARXIV-2209-06899: https://sociolens-web.pages.dev/sources/src-sl-argyle-busby-fulda-gubler-rytting-wingate-arxiv-2209-06899/v1.0/}
}