publications | Polina Tsvilodub

2025

Relevant answers to polar questions

Robert D Hawkins, Polina Tsvilodub, Claire Augusta Bergey, and 2 more authors

Philosophical Transactions B, 2025

@article{hawkins2025relevant,
  title = {Relevant answers to polar questions},
  author = {Hawkins, Robert D and Tsvilodub, Polina and Bergey, Claire Augusta and Goodman, Noah D and Franke, Michael},
  journal = {Philosophical Transactions B},
  volume = {380},
  number = {1932},
  pages = {20230505},
  year = {2025},
  publisher = {The Royal Society},
  url = {https://osf.io/preprints/psyarxiv/8tnfd_v1}
}

Integrating Neural and Symbolic Components in a Model of Pragmatic Question-Answering

Polina Tsvilodub, Robert D. Hawkins, and Michael Franke

Society for Computation in Linguistics, 2025

Bib

@article{tsvilodub2025integrating,
  title = {Integrating Neural and Symbolic Components in a Model of Pragmatic Question-Answering},
  author = {Tsvilodub, Polina and Hawkins, Robert D. and Franke, Michael},
  journal = {Society for Computation in Linguistics},
  volume = {8},
  number = {1},
  year = {2025},
  publisher = {University of Massachusetts Amherst Libraries}
}

Non-literal understanding of number words by language models

Polina Tsvilodub, Kanishk Gandhi, Haoran Zhao, and 3 more authors

In Proceedings of the annual meeting of the Cognitive Science Society, 2025

Bib

@inproceedings{tsvilodub2025non,
  title = {Non-literal understanding of number words by language models},
  author = {Tsvilodub, Polina and Gandhi, Kanishk and Zhao, Haoran and Fr{\"a}nken, Jan-Philipp and Franke, Michael and Goodman, Noah D},
  booktitle = {Proceedings of the annual meeting of the Cognitive Science Society},
  volume = {47},
  year = {2025},
  url = {https://escholarship.org/uc/item/08c3s3jt}
}

2024

Predictions from language models for multiple-choice tasks are not robust under variation of scoring methods

Polina Tsvilodub, Hening Wang, Sharon Grosch, and 1 more author

arXiv preprint arXiv:2403.00998, 2024

Bib

@article{tsvilodub2024predictions,
  title = {Predictions from language models for multiple-choice tasks are not robust under variation of scoring methods},
  author = {Tsvilodub, Polina and Wang, Hening and Grosch, Sharon and Franke, Michael},
  journal = {arXiv preprint arXiv:2403.00998},
  year = {2024}
}

Experimental Pragmatics with Machines: Testing LLM Predictions for the Inferences of Plain and Embedded Disjunctions

Polina Tsvilodub, Paul Marty, Sonia Ramotowska, and 2 more authors

In Proceedings of the annual meeting of the Cognitive Science Society, 2024

Bib

@inproceedings{tsvilodub2024experimental,
  title = {Experimental Pragmatics with Machines: Testing {LLM} Predictions for the Inferences of Plain and Embedded Disjunctions},
  author = {Tsvilodub, Polina and Marty, Paul and Ramotowska, Sonia and Romoli, Jacopo and Franke, Michael},
  booktitle = {Proceedings of the annual meeting of the Cognitive Science Society},
  volume = {46},
  year = {2024}
}

Bayesian Statistical Modeling with Predictors from LLMs

Michael Franke, Polina Tsvilodub, and Fausto Carcassi

arXiv preprint arXiv:2406.09012, 2024

Bib

@article{franke2024bayesian,
  title = {Bayesian Statistical Modeling with Predictors from LLMs},
  author = {Franke, Michael and Tsvilodub, Polina and Carcassi, Fausto},
  journal = {arXiv preprint arXiv:2406.09012},
  year = {2024}
}

Cognitive Modeling with Scaffolded LLMs: A Case Study of Referential Expression Generation

Polina Tsvilodub, Michael Franke, and Fausto Carcassi

In ICML 2024 Workshop on LLMs and Cognition, 2024

Bib

@inproceedings{tsvilodub2024cognitive,
  title = {Cognitive Modeling with Scaffolded {LLM}s: A Case Study of Referential Expression Generation},
  author = {Tsvilodub, Polina and Franke, Michael and Carcassi, Fausto},
  booktitle = {ICML 2024 Workshop on LLMs and Cognition},
  year = {2024},
  url = {https://openreview.net/forum?id=gnGhkVfhje}
}

2023

Evaluating pragmatic abilities of image captioners on A3DS

Polina Tsvilodub and Michael Franke

In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), Jul 2023

Abs DOI Bib

Evaluating grounded neural language model performance with respect to pragmatic qualities like the trade off between truthfulness, contrastivity and overinformativity of generated utterances remains a challenge in absence of data collected from humans. To enable such evaluation, we present a novel open source image-text dataset “Annotated 3D Shapes” (A3DS) comprising over nine million exhaustive natural language annotations and over 12 million variable-granularity captions for the 480,000 images provided by Burgess & Kim (2018).We showcase the evaluation of pragmatic abilities developed by a task-neutral image captioner fine-tuned in a multi-agent communication setting to produce contrastive captions. The evaluation is enabled by the dataset because the exhaustive annotations allow to quantify the presence of contrastive features in the model’s generations. We show that the model develops human-like patterns (informativity, brevity, over-informativity for specific features (e.g., shape, color biases)).
@inproceedings{tsvilodub-franke-2023-evaluating, title = {Evaluating pragmatic abilities of image captioners on {A}3{DS}}, author = {Tsvilodub, Polina and Franke, Michael}, editor = {Rogers, Anna and Boyd-Graber, Jordan and Okazaki, Naoaki}, booktitle = {Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)}, month = jul, year = {2023}, address = {Toronto, Canada}, publisher = {Association for Computational Linguistics}, url = {https://aclanthology.org/2023.acl-short.110/}, doi = {10.18653/v1/2023.acl-short.110}, pages = {1277--1285} }

Overinformative Question Answering by Humans and Machines

Polina Tsvilodub, Michael Franke, Robert Hawkins, and 1 more author

In Proceedings of the 45th Annual Conference of the Cognitive Science Society, Jul 2023

Bib

@inproceedings{tsvilodub2023overinformative,
  title = {Overinformative Question Answering by Humans and Machines},
  author = {Tsvilodub, Polina and Franke, Michael and Hawkins, Robert and Goodman, Noah D.},
  booktitle = {Proceedings of the 45th Annual Conference of the Cognitive Science Society},
  year = {2023},
  organization = {Cognitive Science Society}
}

The role of relevance, competence, and priors for scalar inferences

Polina Tsvilodub, Bob Tiel, and Michael Franke

Experiments in Linguistic Meaning, Jul 2023

Bib

@article{tsvilodub2023role,
  title = {The role of relevance, competence, and priors for scalar inferences},
  author = {Tsvilodub, Polina and van Tiel, Bob and Franke, Michael},
  journal = {Experiments in Linguistic Meaning},
  volume = {2},
  pages = {288--298},
  year = {2023}
}

2022

Improving a gamified language learning chatbot through AI and UX boosting

Polina Tsvilodub, Esther Chevalier, Vera Klütz, and 3 more authors

In The Learning Ideas Conference, Jul 2022

Bib

@inproceedings{tsvilodub2022improving,
  title = {Improving a gamified language learning chatbot through {AI} and {UX} boosting},
  author = {Tsvilodub, Polina and Chevalier, Esther and Kl{\"u}tz, Vera and Oberbeck, Tobias and Sigetova, Kristina and Wollatz, Frederik},
  booktitle = {The Learning Ideas Conference},
  pages = {557--569},
  year = {2022},
  organization = {Springer}
}

2021

Modeling German Word Order Acquisition via Bayesian Inference

Annika Heuser and Polina Tsvilodub

In Proceedings of the Society for Computation in Linguistics 2021, Feb 2021

Bib

@inproceedings{heuser-tsvilodub-2021-modeling,
  title = {Modeling {G}erman Word Order Acquisition via {B}ayesian Inference},
  author = {Heuser, Annika and Tsvilodub, Polina},
  editor = {Ettinger, Allyson and Pavlick, Ellie and Prickett, Brandon},
  booktitle = {Proceedings of the {S}ociety for {C}omputation in {L}inguistics 2021},
  month = feb,
  year = {2021},
  address = {Online},
  publisher = {Association for Computational Linguistics},
  url = {https://aclanthology.org/2021.scil-1.49/},
  pages = {420--424}
}

Comparing Symbolic Models of Language via Bayesian Inference (Student Abstract)

Annika Heuser and Polina Tsvilodub

In Proceedings of the AAAI Conference on Artificial Intelligence, Feb 2021

Bib

@inproceedings{heuser2021comparing,
  title = {Comparing Symbolic Models of Language via {B}ayesian Inference (Student Abstract)},
  author = {Heuser, Annika and Tsvilodub, Polina},
  booktitle = {Proceedings of the AAAI Conference on Artificial Intelligence},
  volume = {35},
  number = {18},
  pages = {15799--15800},
  year = {2021}
}

2020

Informational goals, sentence structure, and comparison class inference

Michael Henry Tessler, Polina Tsvilodub, Jesse Snedeker, and 1 more author

In Proceedings of the annual meeting of the Cognitive Science Society, Feb 2020

Bib

@inproceedings{tessler2020informational,
  title = {Informational goals, sentence structure, and comparison class inference},
  author = {Tessler, Michael Henry and Tsvilodub, Polina and Snedeker, Jesse and Levy, Roger P},
  booktitle = {Proceedings of the annual meeting of the Cognitive Science Society},
  volume = {42},
  year = {2020}
}