Elamur Research⌕
☰
home / papers / 2026 / Exploring Cognitively Informed Sentence Simplification with Gaze-Guided Text Generation

Exploring Cognitively Informed Sentence Simplification with Gaze-Guided Text Generation

Summary

Text simplification systems usually learn what counts as "easy" from human judgments or crowdsourced reference simplifications. These can diverge from how hard a text actually is to process. This paper tests whether eye-tracking data can steer simplification directly, using gaze-guided text generation (GGTG). In GGTG, a gaze model reranks the next-token candidates proposed by a language model, and a tunable gaze weight sets how strongly predicted reading effort is pushed down. The authors fine-tuned GPT-2 large to predict five word-level reading measures from four English naturalistic-reading corpora. They also built "residual" variants that factor out word length and frequency, so the models would focus on deeper sources of difficulty. Generation used Llama 3.2 (3B, instruction-tuned) with either a paraphrase or a simplify prompt, and the system was evaluated on ASSET. Mild gaze steering slightly improved automatic metrics, even on top of an explicit simplify prompt. However, the changes stayed mostly lexical, and stronger steering produced longer outputs with deeper syntax. The work matters as a first attempt to put cognitive signals inside the simplification process itself, and because the method is modular it can be combined with prompting or fine-tuning.

Key points

  • Gaze models: GPT-2 large (774M) was trained separately for each of five measures:
    • duration of the initial fixation on a word
    • total time spent on a word during the first pass
    • time from first reaching a word until moving beyond it to the right
    • first-pass skipping rate
    • first-pass regression rate
  • Training data: EMTeC, OneStopQA, MECO-L1 English and Provo, with a 90/10 split by text. Residual models predict what remains after a linear regression on word length and Zipf frequency (wordfreq).
  • Evaluation setup: ASSET validation set (2,000 sentences, ten references each), beam search with beam size 8, and gaze weights from 0 to −3. Metrics were SARI, LENS, LENS-SALSA and FKGL, plus word frequency and Stanza dependency tree depth.
  • Results by gaze weight: At weight −1, SARI and LENS-SALSA rose slightly, but they fell at stronger weights. Around −3, longer sentences cancelled the FKGL gains, and tree depth tended to increase.
  • Results by measure: First-pass reading time gave the most consistent improvements. Residual models had weaker effects overall. Code is available at github.com/mainlp/gaze-guided-sentence-simplification.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Machine translation · Natural language processing

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.63317/3cju9gvduaga

@inproceedings{sauberli2026exploring,
  title = {{Exploring Cognitively Informed Sentence Simplification with Gaze-Guided Text Generation}},
  author = {Säuberli, Andreas and Frassinelli, Diego and Plank, Barbara},
  booktitle = {Proceedings fo the Second International Workshop on Eye-Tracking Resources and Evaluation for Human-Aligned NLP},
  year = {2026},
  doi = {10.63317/3cju9gvduaga},
  url = {https://doi.org/10.63317/3cju9gvduaga}
}

References

43 references, 7 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Sweta Agrawal and Marine Carpuat. 2023. Controlling pre-trained language models for grade-specific text simplification. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 12807–12819. Association for Computational Linguistics. ↗
  2. Oliver Alonzo, Jessica Trussell, Becca Dingman, and Matt Huenerfauth. 2021. Comparison of methods for evaluating complexity of simplified texts among deaf and hard-of-hearing adults at different literacy levels. In Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems, CHI ’21, pages 1–12. ACM. find
  3. Fernando Alva-Manchego, Louis Martin, Antoine Bordes, Carolina Scarton, Benoît Sagot, and Lucia Specia. 2020. ASSET: A dataset for tuning and evaluation of sentence simplification models with multiple rewriting transformations. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 4668–4679. Association for Computational Linguistics. find our card
  4. Fernando Alva-Manchego, Carolina Scarton, and Lucia Specia. 2021. The (un)suitability of automatic evaluation metrics for text simplification. Computational Linguistics, 47(4):861–889. ↗ our card
  5. Yevgeni Berzak, Jonathan Malmaud, Omer Shubi, Yoav Meiri, Ella Lion, and Roger Levy. 2025. OneStop: A 360-participant English eye tracking dataset with different reading regimes. Scientific Data, 12(1). ↗
  6. Lena S. Bolliger, Patrick Haller, Isabelle C. R. Cretton, David R. Reich, Tannon Kew, and Lena A. Jäger. 2025. EMTeC: A corpus of eye movements on machine-generated texts. Behavior Research Methods, 57(7). ↗
  7. Luisa Carrer, Andreas Säuberli, Martin Kappus, and Sarah Ebling. 2024. Towards holistic human evaluation of automatic text simplification. In Proceedings of the Fourth Workshop on Human Evaluation of NLP Systems (HumEval) @ LREC-COLING 2024, pages 71–80, Torino, Italia. ELRA and ICCL. ↗
  8. William Coster and David Kauchak. 2011. Simple English Wikipedia: A new text simplification task. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, pages 665–669, Portland, Oregon, USA. Association for Computational Linguistics. find our card
  9. Aline Godfroid. 2019. Eye Tracking in Second Language Acquisition and Bilingualism: A Research Synthesis and Methodological Guide. Routledge. ↗
  10. Natalia Grabar and Horacio Saggion. 2022. Evaluation of automatic text simplification: Where are we now, where should we go from here. In Actes de la 29e Conférence sur le Traitement Automatique des Langues Naturelles. Volume 1 : conférence principale, pages 453–463, Avignon, France. ATALA. find
  11. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, Amy Yang, Angela Fan, Anirudh Goyal, Anthony Hartshorn, Aobo Yang, Archi Mitra, Archie Sravankumar, Artem Korenev, Arthur Hinsvark, Arun Rao, et al. 2024. The Llama 3 herd of models. arXiv. find
  12. Keren Gruteke Klein, Shachar Frenkel, Omer Shubi, and Yevgeni Berzak. 2025a. Surprisal takes it all: Eye tracking based cognitive evaluation of text readability measures. arXiv. find
  13. Keren Gruteke Klein, Omer Shubi, Shachar Frenkel, and Yevgeni Berzak. 2025b. The effect of text simplification on reading fluency and reading comprehension in L1 English speakers. OSF. find
  14. David Heineman, Yao Dou, Mounica Maddela, and Wei Xu. 2023. Dancing between success and failure: Edit-level simplification evaluation using SALSA. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 3466–3495. Association for Computational Linguistics. find
  15. Nora Hollenstein, Maria Barrett, and Marina Björnsdóttir. 2022. The copenhagen corpus of eye tracking recordings from natural reading of Danish texts. In Proceedings of the Thirteenth Language Resources and Evaluation Conference, pages 1712–1720, Marseille, France. European Language Resources Association. find
  16. Oksana Ivchenko and Natalia Grabar. 2024. Study of medical text reading and comprehension through eye-tracking fixations. In Proceedings of the First Workshop on Patient-Oriented Language Processing (CL4Health) @ LREC-COLING 2024, pages 84–92, Torino, Italia. ELRA and ICCL. find
  17. Deborah Noemie Jakobi, Maja Stegenwallner-Schütz, Nora Hollenstein, Cui Ding, Ramune Kaspere, Ana Matić Škorić, Eva Pavlinusic Vilus, Stefan Frank, Marie-Luise Müller, Kristine M Jensen de López, Nik Kharlamov, Hanne B. Søndergaard Knudsen, Yevgeni Berzak, Ella Lion, Irina A. Sekerina, Cengiz Acarturk, Mohd Faizan Ansari, Katarzyna Harezlak, Pawel Kasprowski, Ana Bautista, et al. 2025. MultiplEYE: Creating a multilingual eye-tracking-while-reading corpus. In Proceedings of the 2025 Symposium on Eye Tracking Research and Applications, ETRA ’25, pages 1–11. ACM. find
  18. Marcel A. Just and Patricia A. Carpenter. 1980. A theory of reading: From eye fixations to comprehension. Psychological Review, 87(4):329–354. ↗
  19. Tannon Kew and Sarah Ebling. 2022. Target-level sentence simplification as controlled paraphrasing. In Proceedings of the Workshop on Text Simplification, Accessibility, and Readability (TSAR-2022), pages 28–42. Association for Computational Linguistics. ↗
  20. J. Peter Kincaid, Robert P. Fishburne Jr, Richard L Rogers, and Brad S. Chissom. 1975. Derivation of new readability formulas (automated readability index, fog count and Flesch reading ease formula) for navy enlisted personnel. Technical report. ↗
  21. Victor Kuperman, Noam Siegelman, Sascha Schroeder, Cengiz Acartürk, Svetlana Alexeeva, Simona Amenta, Raymond Bertram, Rolando Bonandrini, Marc Brysbaert, Daria Chernova, Sara Maria Da Fonseca, Nicolas Dirix, Wouter Duyck, Argyro Fella, Ram Frost, Carolina A. Gattei, Areti Kalaitzi, Kaidi Lõo, Marco Marelli, Kelly Nisbet, et al. 2023. Text reading in English as a second language: Evidence from the Multilingual Eye-Movements Corpus. Studies in Second Language Acquisition, 45(1):3–37. ↗
  22. Steven G. Luke and Kiel Christianson. 2017. The Provo Corpus: A large eye-tracking corpus with predictability norms. Behavior Research Methods, 50(2):826–833. ↗
  23. Mounica Maddela, Yao Dou, David Heineman, and Wei Xu. 2023. LENS: A learnable evaluation metric for text simplification. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 16383–16408. Association for Computational Linguistics. find our card
  24. Louis Martin, Éric de la Clergerie, Benoît Sagot, and Antoine Bordes. 2020. Controllable sentence simplification. In Proceedings of the Twelfth Language Resources and Evaluation Conference, pages 4689–4698, Marseille, France. European Language Resources Association. ↗
  25. Louis Martin, Angela Fan, Éric de la Clergerie, Antoine Bordes, and Benoît Sagot. 2022. MUSS: Multilingual unsupervised sentence simplification by mining paraphrases. In Proceedings of the Thirteenth Language Resources and Evaluation Conference, pages 1651–1664, Marseille, France. European Language Resources Association. find our card
  26. Daiki Nishihara, Tomoyuki Kajiwara, and Yuki Arase. 2019. Controllable text simplification with lexical constraint loss. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics: Student Research Workshop, pages 260–266. Association for Computational Linguistics. ↗
  27. Gustavo Paetzold and Lucia Specia. 2017. Lexical simplification with neural ranking. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 2, Short Papers, pages 34–40, Valencia, Spain. Association for Computational Linguistics. find
  28. Fabian Pedregosa, Gaël Varoquaux, Alexandre Gramfort, Vincent Michel, Bertrand Thirion, Olivier Grisel, Mathieu Blondel, Peter Prettenhofer, Ron Weiss, Vincent Dubourg, Jake Vanderplas, Alexandre Passos, David Cournapeau, Matthieu Brucher, Matthieu Perrot, and Édouard Duchesnay. 2011. Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12(85):2825–2830. ↗
  29. Peng Qi, Yuhao Zhang, Yuhui Zhang, Jason Bolton, and Christopher D. Manning. 2020. Stanza: A Python natural language processing toolkit for many human languages. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics: System Demonstrations, pages 101–108. Association for Computational Linguistics. find our card
  30. Alec Radford, Jeff Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. 2019. Language models are unsupervised multitask learners. OpenAI. find
  31. Keith Rayner. 1998. Eye movements in reading and information processing: 20 years of research. Psychological Bulletin, 124(3):372–422. ↗
  32. David R. Reich, Shuwen Deng, Marina Björnsdóttir, Lena Jäger, and Nora Hollenstein. 2024. Reading does not equal reading: Comparing, simulating and exploiting reading behavior across populations. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 13586–13594, Torino, Italia. ELRA and ICCL. find
  33. Luz Rello, Ricardo Baeza-Yates, Stefan Bott, and Horacio Saggion. 2013. Simplify or help? Text simplification strategies for people with dyslexia. In Proceedings of the 10th International Cross-Disciplinary Conference on Web Accessibility, W4A ’13, pages 1–10. ACM. find
  34. Carolina Scarton and Lucia Specia. 2018. Learning simplifications for specific target audiences. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 712–718. Association for Computational Linguistics. find
  35. Noam Siegelman, Sascha Schroeder, Cengiz Acartürk, Hee-Don Ahn, Svetlana Alexeeva, Simona Amenta, Raymond Bertram, Rolando Bonandrini, Marc Brysbaert, Daria Chernova, Sara Maria Da Fonseca, Nicolas Dirix, Wouter Duyck, Argyro Fella, Ram Frost, Carolina A. Gattei, Areti Kalaitzi, Nayoung Kwon, Kaidi Lõo, Marco Marelli, et al. 2022. Expanding horizons of cross-linguistic research on reading: The Multilingual Eye-Movement Corpus (MECO). Behavior Research Methods, 54(6):2843–2863. ↗
  36. Noam Siegelman, Sascha Schroeder, Yaqian Borogjoon Bao, Cengiz Acartürk, Niket Agrawal, Lena S. Bolliger, Jan Brasser, César Campos-Rojas, Denis Drieghe, Dušica Filipović Ðurđević, Sofya Goldina, Romualdo Ibáñez Orellana, Lena A. Jäger, Ómar I. Jóhannesson, Anurag Khare, Nik Kharlamov, Hanne B. S. Knudsen, Árni Kristjánsson, Charlotte E. Lee, Jun Ren Lee, et al. 2025. Wave 2 of the Multilingual Eye-Movement Corpus (meco): New text reading data across languages. Scientific Data, 12(1). ↗
  37. Abhinav Deep Singh, Poojan Mehta, Samar Husain, and Rajkumar Rajakrishnan. 2016. Quantifying sentence complexity based on eye-tracking measures. In Proceedings of the Workshop on Computational Linguistics for Linguistic Complexity (CL4LC), pages 202–212, Osaka, Japan. find
  38. The COLING 2016 Organizing Committee. find
  39. Robyn Speer. 2022. wordfreq. Zenodo. find
  40. Andreas Säuberli, Darja Jepifanova, Diego Frassinelli, and Barbara Plank. 2026. Controlling reading ease with gaze-guided text generation. In Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2383–2397. Association for Computational Linguistics. ↗
  41. Sowmya Vajjala, Detmar Meurers, Alexander Eitel, and Katharina Scheiter. 2016. Towards grounding computational linguistic approaches to readability: Modeling reader-text interaction for easy and difficult texts. In Proceedings of the Workshop on Computational Linguistics for Linguistic Complexity (CL4LC), pages 38–48, Osaka, Japan. find
  42. The COLING 2016 Organizing Committee. find
  43. Wei Xu, Courtney Napoles, Ellie Pavlick, Quanze Chen, and Chris Callison-Burch. 2016. Optimizing statistical machine translation for text simplification. Transactions of the Association for Computational Linguistics, 4:401–415. ↗ our card

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/sauberli-2026-exploring-cognitively-informed-sentence-simplification/