Can Large Language Models (or Humans) Disentangle Text?

Nicolas Audinet de Pieuchon | Adel Daoud | Connor Jerzak | Moa Johansson | Richard Johansson

Paper Code Data .bib


Abstract: We investigate the potential of large language models (LLMs) to disentangle text variables—to remove the textual traces of an undesired forbidden variable in a task sometimes known as text distillation and closely related to the fairness in AI and causal inference literature. We employ a range of various LLM approaches in an attempt to disentangle text by identifying and removing information about a target variable while preserving other relevant signals. We show that in the strong test of removing sentiment, the statistical association between the processed text and sentiment is still detectable to machine learning classifiers post-LLM-disentanglement. Furthermore, we find that human annotators also struggle to disentangle sentiment while preserving other semantic content. This suggests there may be limited separability between concept variables in some text contexts, highlighting limitations of methods relying on text-level transformations and also raising questions about the robustness of disentanglement methods that achieve statistical independence in representation space if this is difficult for human coders operating on raw text to attain.

References

Nicolas Audinet de Pieuchon, Adel Daoud, Connor T. Jerzak, Moa Johansson, Richard Johansson. Can Large Language Models (or Humans) Disentangle Text?. Proceedings of the Sixth Workshop on Natural Language Processing and Computational Social Science (NLP+ CSS 2024): 57-67, 2024.
@article{pieuchon2024can,
  title={Can Large Language Models (or Humans) Disentangle Text?},
  author={Pieuchon, Nicolas Audinet de and Adel Daoud and Connor T. Jerzak and Moa Johansson and Richard Johansson},
  journal={Proceedings of the Sixth Workshop on Natural Language Processing and Computational Social Science (NLP+ CSS 2024)},
  year={2024},
  pages={57--67},
  publisher={}
}
[Data][Code]

Related Work

Adel Daoud, Connor T. Jerzak, Richard Johansson. Conceptualizing Treatment Leakage in Text-based Causal Inference. Proceedings of the Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL): 5638-5645, 2022.
@article{daoud2022conceptualizing,
  title={Conceptualizing Treatment Leakage in Text-based Causal Inference},
  author={Daoud, Adel and Connor T. Jerzak and Richard Johansson},
  journal={Proceedings of the Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL)},
  year={2022},
  volume={},
  pages={5638-5645},
  publisher={}
}
[Overview]

Connor T. Jerzak, Gary King, Anton Strezhnev. An Improved Method of Automated Nonparametric Content Analysis for Social Science. Political Analysis, 31(1): 42-58, 2023.
@article{jerzak2023improved,
  title={An Improved Method of Automated Nonparametric Content Analysis for Social Science},
  author={Jerzak, Connor T. and Gary King and Anton Strezhnev},
  journal={Political Analysis},
  year={2023},
  volume={31},
  number={1},
  pages={42-58},
  publisher={}
}
[Overview]

Back to Research
Back to Home