Multilingual Language Models Predict Human Reading Behavior (2021.naacl-main)

Copied to clipboard

Challenge: Recent studies show that cognitively motivated "attention" mechanism in neural models is not a good indicator for relative importance.
Approach: They compare the performance of language-specific and multilingual pretrained transformer models to predict reading time measures reflecting natural human sentence processing.
Outcome: The proposed models predict reading time measures on Dutch, English, German, and Russian texts.

Similar Papers

Scaling in Cognitive Modelling: a Multilingual Approach to Human Reading Times (2023.acl-short)

Copied to clipboard

Challenge: Neural language models provide conditional probability distributions over the lexicon that are predictive of human processing times.
Approach: They propose to use a transformer-based model to generate probabilistic estimates that are less predictive of early eye-tracking measurements reflecting lexical access and early semantic integration.
Outcome: The proposed models show that larger models capture late eye-tracking measurements that reflect the full integration of a word into the current language context.
Measuring the Impact of (Psycho-)Linguistic and Readability Features and Their Spill Over Effects on the Prediction of Eye Movement Patterns (2022.acl-long)

Copied to clipboard

Challenge: Existing work to predict gaze patterns during naturalistic reading has not been conducted on general text characteristics.
Approach: They propose to use two eye-tracking corpora of naturalistic reading and two language models to test their performance.
Outcome: The proposed models predict eye-tracking measures during naturalistic reading and language processing.
Modeling Nonnative Sentence Processing with L2 Language Models (2024.emnlp-main)

Copied to clipboard

Challenge: Experimental results show that while all of the LMs’ word surprisals improve prediction of L2 reading times, there is no reliable effect of the choice of L1’s L1.
Approach: They pretrain GPT2 on 6 different first languages, followed by English as the second language (L2).
Outcome: The pretraining of L1 improves prediction of L2 reading times, but there is no reliable effect of the pretraining L1 on the model's performance on English speakers.
From Human Reading to NLM Understanding: Evaluating the Role of Eye-Tracking Data in Encoder-Based Models (2025.acl-long)

Copied to clipboard

Challenge: integrating eye-tracking features into Neural Language Models does not degrade downstream task performance, enhances alignment between model attention and human attention patterns, and compresses the embedding space.
Approach: They used eye-gaze data from the Ghent Eye-Tracking Corpus to investigate how integrating knowledge of human reading behavior impacts Neural Language Models.
Outcome: The proposed approach does not degrade downstream task performance, enhances alignment between model attention and human attention patterns, and compresses the embedding space.
Every word counts: A multilingual analysis of individual human alignment with model attention (2022.aacl-short)

Copied to clipboard

Challenge: Using eye-tracking data, fixation durations are often not considered in generalisation studies because of individual differences.
Approach: They analyse eye-tracking data from speakers of 13 different languages reading . they find significant differences between languages but also individual reading behaviour .
Outcome: The proposed model can be used to improve the generalization of ML models and allow for more personalized and fair applications.
Transformer-Based Language Model Surprisal Predicts Human Reading Times Best with About Two Billion Training Tokens (2023.findings-emnlp)

Copied to clipboard

Challenge: Recent studies have drawn conflicting conclusions about the relationship between the quality of a language model and the ability of its surprisal estimates to predict human reading times.
Approach: They propose to evaluate surprisal estimates from Transformer-based language model variants that vary systematically in the amount of training data and model capacity on their ability to predict human reading times.
Outcome: The proposed model variants with contemporary model capacities provide the best fit after seeing about two billion training tokens, while smaller models show a ‘tipping point’ at convergence after the decrease in language model perplexity .
Probing for Reading Times (2026.acl-long)

Copied to clipboard

Challenge: a large body of work on probing has demonstrated that language model representations encode a wealth of linguistic information, but it remains unclear whether they also capture cognitive signals about human processing.
Approach: They use regularized linear regression to compare language model representations against scalar predictors.
Outcome: The representations from early layers outperform surprisal in predicting early-pass measures such as first fixation and gaze duration.
Do Transformer Models Show Similar Attention Patterns to Task-Specific Human Gaze? (2022.acl-long)

Copied to clipboard

Challenge: We compare attention functions in pre-trained language models to human eye fixation patterns during task-specific reading tasks.
Approach: They compare attention functions in large-scale pre-trained language models to classical cognitive models of human attention by using a dataset with eye-tracking recordings of native speakers of English.
Outcome: The proposed model is as predictive of human eye fixation patterns as classical cognitive models of human attention.
Cross-Lingual Transfer of Cognitive Processing Complexity (2023.findings-eacl)

Copied to clipboard

Challenge: Recent studies indicate that multilingual language models utilize structural similarities between languages to facilitate cross-lingual transfer.
Approach: They propose a multilingual model that uses structural similarities between languages to facilitate cross-lingual transfer by a meaningful bias towards sentence length and cross-linguistic differences.
Outcome: The proposed model can predict varied patterns for 13 languages, despite being fine-tuned only on English data.
Assessing the Syntactic Capabilities of Transformer-based Multilingual Language Models (2021.findings-acl)

Copied to clipboard

Challenge: Multilingual Transformer-based language models have been shown to be excellent learners in crosslingual transfer tasks.
Approach: They evaluate the syntactic generalization capabilities of BERT and RoBERTa models on English and Spanish tests.
Outcome: The proposed models perform well on English and Spanish tests, and the proposed tests are compared against models on the same language and models on two different languages.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations