Pith. sign in

REVIEW 4 major objections 5 minor 102 references

Gaze Behavior in Visual World Experiments Can be Modeled With Off-the-shelf Language-Vision Encoders

T0 review · 4 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read No fine-tuning: CLIP attributions replicate predictive gaze in visual-world experiments.

desk verdict Genuinely novel zero-shot method for visual-world gaze, with a plausible qualitative replication that is unfortunately dependent on a post hoc normalization choice; worth serious engagement, but the robustness claim needs revision. read the letter →

arxiv 2608.07282 v1 pith:AFVCIMKU submitted 2026-08-07 cs.CL cs.CV

Rahul Murali Shankar , Titus von der Malsburg , Sebastian Padó This is my paper · ORCID
classification cs.CLcs.CV
keywords visualworldparadigmCLIPIntegratedJacobianspredictiveprocessinggazemodelingattributionmethodsvision-languagemodelspsycholinguistics
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper claims that off-the-shelf CLIP vision-language encoders, paired with an attribution method called Integrated Jacobians, can reproduce the anticipatory gaze pattern observed in human visual-world experiments. When the input contains a constraining verb such as "eat," the model assigns higher attribution to the target object (the cake) before that object is named, and this advantage disappears once the object is mentioned. The effect holds across four CLIP models without any fine-tuning or generative decoding, suggesting that predictive gaze behavior can emerge from correlational multimodal similarity rather than dedicated prediction machinery.

What carries the argument

Integrated Jacobians, a generalization of Integrated Gradients to dual encoders, decompose CLIP's similarity score between a caption and an image into attribution scores for each token and image patch. ReLU normalization clamps negative attributions to zero so the scores behave like fixation saliencies, and bounding-box aggregation sums patch scores into per-object probabilities with a one-patch margin. Crucially, the encoder sees only each prefix of the linguistic input in isolation, preventing the attribution from using the sentence-final object as context.

What would settle it

Record human eye movements on a second set of visual-world stimuli with the same Altmann-Kamide design, run the identical attribution pipeline on those stimuli, and check whether the item-level verb-type by region interaction in CLIP attributions predicts the item-level interaction in human target preference; if the model interaction does not track the human interaction across items, the linking hypothesis fails.

Watch

Extended reading notes

Core claim

The paper's central claim is that ReLU-normalized Integrated Jacobian attributions from CLIP models qualitatively replicate the time-course of human predictive processing in the Altmann-Kamide stimuli: a constraining verb reliably drives stronger target-object preference in the pre-object region, and this advantage attenuates once the object noun is encountered. Across the four models, the population-level interaction between verb type and region is significant and negative, confirming that the predictive advantage is localized to the window before the object is named.

Load-bearing premise

The claim stands on the assumption that ReLU-normalized, bounding-box-aggregated CLIP attribution scores can stand in for human fixation probabilities; the paper calls this "arguably sufficient" for simple static displays and does not validate it against recorded eye movements.

Editorial extensions

If this is right

  • The qualitative time course of human predictive gaze can be reproduced with no training, no fine-tuning, and no generative decoder.
  • Because the CLIP language encoder is non-incremental and sees each prefix separately, the predictive signal is attributable to the constraining verb's content rather than to incremental processing dynamics.
  • Larger, higher-accuracy CLIP models tended to show larger pre-object effects in the four models tested, though the paper treats this pattern as hypothesis-generating given the small number of checkpoints.
  • The prediction step itself needs no object labels or bounding boxes; bounding boxes are used only for evaluation aggregation, so the approach avoids manual object categorization during prediction.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A direct quantitative comparison against recorded fixation proportions on the same or new stimuli would be a stronger test than the qualitative interaction reported here; the paper's linking hypothesis does not yet license precise probability predictions.
  • The same pipeline could be pointed at other visual-world phenomena, such as semantic competitors or color-based preactivation, to test whether CLIP attributions capture featural anticipation and not just selectional restrictions of verbs.
  • Varying the one-patch bounding-box margin and comparing against human-defined object regions would show how much of the reported interaction depends on the annotation choice, which the normalization ablation does not cover.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes a method to model gaze in visual world experiments without any fine-tuning or human gaze training data. It encodes each image with each linguistic prefix of the sentence using off-the-shelf CLIP bi-encoders, computes Integrated Jacobians attributions between text tokens and image patches, applies a ReLU normalization to the per-patch attributions, and aggregates them over manually annotated bounding boxes to obtain a target-preference score. The authors test the approach on the 18 stimuli of Altmann and Kamide (1999) with four CLIP checkpoints and report that a constraining verb induces a significantly stronger target-object preference before the noun is encountered than a non-constraining verb, and that this advantage attenuates at the noun, mirroring human anticipatory gaze. They interpret this as evidence that predictive processing can emerge from correlational structure in contrastively trained encoders without a generative or autoregressive objective.

Significance. If the result holds, it is a valuable contribution to computational psycholinguistics: it shows that a simple, non-generative, off-the-shelf vision-language encoder can qualitatively reproduce a seminal anticipatory eye-movement effect without being trained for incremental language processing. The paper has several genuine strengths: the prefix-presentation design in Section 3.6 prevents the encoder from using future context, which is a clean solution to a real leakage problem; the evaluation uses multiple CLIP checkpoints and standard mixed-effects analyses; the failure-mode analysis in Section 5 is honest and informative; and the authors state that code and materials will be released. The central empirical claim, however, rests on two analytic choices that are not robustly justified: the post hoc selection of ReLU normalization and the exclusion of one checkpoint based on its outcome. These issues are fixable with additional sensitivity analyses or pre-specified decision rules, and the linking hypothesis needs either validation against human gaze data or a more careful scoping of the claim.

major comments (4)
  1. [Section 3.5 / Appendix B, Table 3] The choice of ReLU normalization is load-bearing but was made after inspecting the results: the text states that ReLU "turned out to be the best choice," and Table 3 shows that the pre-object constraining-verb effect appears in all four models only under ReLU (three models under Shift, one under Softmax, and none under no normalization). The population-level interaction reported in Section 5 (β = −0.235, p = 0.005) is computed exclusively on ReLU-normalized attributions, so the "across models" claim is not robust to an analytic choice that lacks an a priori theoretical motivation. Please either justify ReLU as the linking function on independent grounds, or report the global mixed-effects model under all four normalization strategies and discuss the resulting sensitivity.
  2. [Footnote 6 / Section 4.2] The exclusion of the fifth checkpoint (OpenCLIP ViT-B/16, dfn2b) is based on the outcome of interest: its attributions are described as degenerate because the target object receives negative attribution at verb onset. This is a selection rule that can only make the reported pattern easier to obtain. Please report the results with this checkpoint included under the same pipeline, or motivate the exclusion with a pre-specified, outcome-independent quality criterion such as a failure of the attribution method to satisfy a basic sanity check.
  3. [Section 3.2 / Section 6, Limitations] The paper claims that normalized Integrated Jacobians attributions "are directly linearly related to human fixations" (Section 1) and that the resulting probabilities "can be interpreted as fixation saliencies" (Section 3.1), but the linking hypothesis is not validated against any human gaze data. For the simple static materials this may be "arguably sufficient," as Section 3.2 says, but the headline claim that gaze behavior "can be modeled" is stronger than what is demonstrated. I recommend either validating the mapping against empirical fixation proportions (e.g., from the original Altmann and Kamide data or a comparable visual world study) or explicitly scoping the claim to attribution-derived target preference and supporting the linking assumption with evidence from prior work.
  4. [Section 3.5 / Section 4.3] The dependent variable depends directly on manually annotated bounding boxes plus a one-patch margin, but the paper reports no reliability assessment or sensitivity analysis for this step. A one-patch change in box boundaries can change the aggregated target preference and therefore the statistical results. Please provide annotation details, inter-annotator agreement if applicable, or a robustness check that varies the margin size and shows that the qualitative conclusions are unchanged.
minor comments (5)
  1. [Throughout] The capitalization of ReLU is inconsistent ("ReLU" in the text and figures, "ReLu" in Section 3.5 and Appendix B); please standardize.
  2. [Figure 1 and Section 1] The caption of Figure 1 describes the panels as attribution heatmaps, while the text in Section 1 says the colored overlay indicates fixation probabilities; please clarify that the figure shows model-derived attributions, not human data.
  3. [Section 6] Section 6 cites Möller et al. (2024) for the finding that CLIP models identify object categories, but Section 3.2 attributes this finding to Möller et al. (2025); please check which reference is intended and make the citations consistent.
  4. [Section 5] There is a typographical error in the model size discussion: "VIT-B/16" should be "ViT-B/16."
  5. [Section 4.1] The sentence "This dataset consists of 18 images which is considered as dataset size with sufficient statistical power for its experimental design" is awkward; consider rephrasing to state that the original study used 18 items and that the authors follow that design.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the target gaze pattern is an external benchmark, not an input to the model or the normalization.

full rationale

The paper's derivation chain is self-contained: it takes off-the-shelf CLIP encoders, applies Integrated Jacobians (a published attribution method), normalizes with ReLU, aggregates to bounding boxes, and compares the resulting target preference to the qualitative human pattern of Altmann and Kamide (1999). No human gaze data are used to fit any parameter, and the CLIP models are not fine-tuned on the experimental materials. The core interaction (verb by region) is an empirical property of the ReLU-normalized attributions, not a consequence of the definitions: target preference is defined from the annotated object boxes and the caption noun, not from fixation data. The paper's self-citations (Moeller et al. 2024 for Integrated Jacobians; Moeller et al. 2025 for CLIP object-category knowledge) provide the attribution tool and motivation, but they do not encode the Altmann-Kamide result; the experiment is an external benchmark. The post hoc selection of ReLU among four normalizations (Appendix B) is a transparency and robustness concern rather than a circular step, since the chosen transform does not define the predicted effect and the alternatives are disclosed. Therefore, no load-bearing reduction to the inputs is present.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central pipeline rests on three unproved premises: the linking hypothesis that attribution scores stand in for fixation probabilities, the faithfulness of Integrated Jacobians as a decomposition of CLIP similarity, and the correctness of the manually annotated bounding boxes. The ReLU normalization and the one-patch margin are hand-chosen analysis decisions rather than derived quantities; the ReLU choice was made post hoc after comparing four normalizations. No new entities are postulated. The paper contributes a new application of existing models and methods, so the axiom burden is moderate.

free parameters (2)
  • Bounding box margin = 1 patch
    Attributions extending beyond objects are captured by adding one patch in all directions; this value is chosen by hand and not varied in sensitivity analysis (Section 3.5).
  • ReLU normalization choice = max(0, a0)
    Selected post hoc because it gave the best significance profile across models (Section 3.5, Appendix B); other normalizations (None, Shift, Softmax) produce weaker or null effects.
assumptions (4)
  • domain assumption Fixation probabilities in visual world experiments can be approximated by CLIP attribution scores to image regions.
    Section 3.2 states the linking hypothesis: for simple static materials it is "arguably sufficient" to assume that fixations reflect the processing of the linguistic input and that positive attributions correspond to relevant objects. If false, the whole mapping from model scores to gaze is invalid.
  • domain assumption Integrated Jacobians faithfully decompose CLIP similarity into token-patch attribution scores.
    Section 3.4 relies on the method of Möller et al. (2024) to compute pairwise attributions; the paper does not independently validate that these scores capture the model's true input-output relationship beyond the cited reference.
  • ad hoc to paper ReLU-clamped attributions form a valid scale of fixation saliency.
    Section 3.5 justifies ReLU as "one of the easiest methods" and Appendix B reveals it was chosen because it produced the strongest effects; this is an ad hoc modeling choice rather than a derived property.
  • domain assumption The original Altmann and Kamide (1999) stimuli and target objects are correctly reproduced and annotated.
    Section 4.1 relies on the original 18 images and captions plus manually annotated bounding boxes; errors in annotation would directly affect object-level attribution aggregation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Gaze Behavior in Visual World Experiments Can be Modeled With Off-the-shelf Language-Vision Encoders." pith.science (2026). https://pith.science/paper/AFVCIMKU

@misc{pith2026260807282,
  author       = {Pith},
  title        = {Pith review of: Gaze Behavior in Visual World Experiments Can be Modeled With Off-the-shelf Language-Vision Encoders},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AFVCIMKU}},
  note         = {Machine review of arXiv:2608.07282}
}
read the original abstract

The recent advances in neural language models have also spurred much work in computational psycholinguistics, asking whether neural LMs are also promising models of human language processing. However, work has been overwhelmingly focused on the unimodal case of written or spoken language. In contrast, multimodal experimental paradigms, like visual world studies that present participants with both visual and linguistic input simultaneously, have been neglected. In this paper, we present a novel approach that predicts gaze behavior in visual world studies. It does so by combining a simple multi-modal bi-encoder model of the CLIP family with a bimodal attribution method. We demonstrate the ability of this approach to robustly replicate the results of a seminal English visual world study which shows hu- man predictive processing. Remarkably, it does so without a generative architecture and without the need for fine-tuning, despite not being trained for this task.

Figures

Figures reproduced from arXiv: 2608.07282 by the authors.

Figure 1
Figure 1. Example attribution heatmaps (yellow/red: positive attribution, green: neutral, blue: negative attribution) [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 3
Figure 3. Target preference (attribution to target object [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Heatmaps for sentence 4 with constraining verb: Failure to understand event [PITH_FULL_IMAGE:figures/full_fig_p012_4.png] view at source ↗
Figures from the paper (1 more)
Figure 5
Figure 5. Figure 5: Heatmaps for sentence 18: Failure to identify target object [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

102 extracted references · 66 canonical work pages

  1. [1]

    Gerry T. M. Altmann and Yuki Kamide , title =. Cognition , url =

  2. [2]

    Explaining Caption-Image Interactions in CLIP Models with Second-Order Attributions

    Möller, Lucas and Tilli, Pascal and Vu, Ngoc Thang and Padó, Sebastian , title =. Transactions on Machine Learning Research , year = 2025, added-at =. arXiv , biburl =:2408.14153 , interhash =

    work page Pith review arXiv 2025

  3. [3]

    Interpretable Text Embeddings and Text Similarity Explanation: A Survey , booktitle =

    Opitz, Juri and Möller, Lucas and Michail, Andrianos and Pad. Interpretable Text Embeddings and Text Similarity Explanation: A Survey , booktitle =. 2025 , editor =. doi:10.18653/v1/2025.emnlp-main.1135 , ISBN =

    doi:10.18653/v1/2025.emnlp-main.1135 2025

  4. [4]

    Proceedings of the 34th International Conference on Machine Learning , year =

    Mukund Sundararajan and Ankur Taly and Qiqi Yan , title =. Proceedings of the 34th International Conference on Machine Learning , year =

  5. [5]

    Approximate Attributions for Off-the-Shelf

    Möller, Lucas and Nikolaev, Dmitry and Pad. Approximate Attributions for Off-the-Shelf. Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics , year =. doi:10.18653/v1/2024.eacl-long.125 , url =

    work page doi:10.18653/v1/2024.eacl-long.125

  6. [6]

    Proceedings of ICLR , year =

    Chankyu Lee and Rajarshi Roy and Mengyao Xu and Jonathan Raiman and Mohammad Shoeybi and Bryan Catanzaro and Wei Ping , title =. Proceedings of ICLR , year =

  7. [7]

    Retrieval-Augmented Generation for Knowledge-Intensive

    Lewis, Patrick and Perez, Ethan and Piktus, Aleksandra and Petroni, Fabio and Karpukhin, Vladimir and Goyal, Naman and K\". Retrieval-Augmented Generation for Knowledge-Intensive. Advances in Neural Information Processing Systems , year =

  8. [8]

    Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , year =

    Plenz, Moritz and Opitz, Juri and Heinisch, Philipp and Cimiano, Philipp and Frank, Anette , title =. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , year =

Show all 102 references
  1. [9]

    Proceedings of the 31st International Conference on Computational Linguistics , year =

    Chollampatt, Shamil and Pham, Minh Quang and Indurthi, Sathish Reddy and Turchi, Marco , title =. Proceedings of the 31st International Conference on Computational Linguistics , year =

  2. [10]

    Computers and Education: Artificial Intelligence , year =

    Rawaa Alatrash and Mohamed Amine Chatti and Qurat Ul Ain and Yipeng Fang and Shoeb Joarder and Clara Siepmann , title =. Computers and Education: Artificial Intelligence , year =. doi:https://doi.org/10.1016/j.caeai.2023.100193 , url =

    2023

  3. [11]

    Reimers, Nils and Gurevych, Iryna , title =. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) , year =

    2019

  4. [12]

    and Fekete, J.-D

    Ghoniem, M. and Fekete, J.-D. and Castagliola, P. , title =. IEEE Symposium on Information Visualization , year =

  5. [13]

    The American Statistician , year =

    Leland Wilkinson and Michael Friendly , title =. The American Statistician , year =

  6. [14]

    Computer Graphics Forum , year =

    Behrisch, Michael and Bach, Benjamin and Henry Riche, Nathalie and Schreck, Tobias and Fekete, Jean-Daniel , title =. Computer Graphics Forum , year =. doi:https://doi.org/10.1111/cgf.12935 , url =

    doi

  7. [15]

    , title =

    Yi, Ji Soo and Kang, Youn ah and Stasko, John and Jacko, J.A. , title =. IEEE Transactions on Visualization and Computer Graphics , year =

  8. [16]

    IEEE Transactions on Visualization and Computer Graphics , year =

    Willett, Wesley and Heer, Jeffrey and Agrawala, Maneesh , title =. IEEE Transactions on Visualization and Computer Graphics , year =

  9. [17]

    and McColeman, Caitlyn M

    Ceja, Cristina R. and McColeman, Caitlyn M. and Xiong, Cindy and Franconeri, Steven L. , title =. IEEE Transactions on Visualization and Computer Graphics , year =. doi:10.1109/TVCG.2020.3030422 , keywords =

    2020

  10. [18]

    , title =

    Roberts, Jonathan C. , title =. Fifth International Conference on Coordinated and Multiple Views in Exploratory Visualization (CMV 2007) , year =

    2007

  11. [19]

    and Woodruff, Allison and Kuchinsky, Allan , title =

    Wang Baldonado, Michelle Q. and Woodruff, Allison and Kuchinsky, Allan , title =. Proceedings of the Working Conference on Advanced Visual Interfaces , year =. doi:10.1145/345513.345271 , numpages =

  12. [20]

    Neo: Generalizing Confusion Matrix Visualization to Hierarchical and Multi-Output Labels , booktitle =

    G\". Neo: Generalizing Confusion Matrix Visualization to Hierarchical and Multi-Output Labels , booktitle =. 2022 , series =. doi:10.1145/3491102.3501823 , articleno =

    2022

  13. [21]

    Silverman, Bernard W , title =

  14. [22]

    IEEE Transactions on Visualization and Computer Graphics , year =

    Bostock, Michael and Ogievetsky, Vadim and Heer, Jeffrey , title =. IEEE Transactions on Visualization and Computer Graphics , year =

  15. [23]

    2020 , doi =

    Honnibal, Matthew and Montani, Ines and Van Landeghem, Sofie and Boyd, Adriane and others , title =. 2020 , doi =

    2020

  16. [24]

    Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: System Demonstrations , year =

    Boyle, Alan David and Cheng, Furui and Zouhar, Vil. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: System Demonstrations , year =. doi:10.18653/v1/2025.emnlp-demos.32 , ISBN =

    2025 doi

  17. [25]

    2018 IEEE Pacific Visualization Symposium (PacificVis) , year =

    Nie, Shaoliang and Healey, Christopher and Padia, Kalpesh and Leeman-Munk, Samuel and Benson, Jordan and Caira, Dave and Sethi, Saratendu and Devarajan, Ravi , title =. 2018 IEEE Pacific Visualization Symposium (PacificVis) , year =

    2018

  18. [26]

    Wagers and Ellen F

    Matthew W. Wagers and Ellen F. Lau and Colin Phillips , title =. Journal of Memory and Language , year =. doi:10.1016/j.jml.2009.04.002 , keywords =

    2009 doi

  19. [27]

    Proceedings of ACL , year =

    Lesci, Pietro and Meister, Clara and Hofmann, Thomas and Vlachos, Andreas and Pimentel, Tiago , title =. Proceedings of ACL , year =

  20. [28]

    Transactions of the Association for Computational Linguistics , year =

    Oh, Byung-Doh and Schuler, William , title =. Transactions of the Association for Computational Linguistics , year =

  21. [29]

    Open Mind , year =

    Cuskley, Christine and Woods, Rebecca and Flaherty, Molly , title =. Open Mind , year =

  22. [30]

    Pearlmutter and Susan M

    Neal J. Pearlmutter and Susan M. Garnsey and Kathryn Bock , title =. Journal of Memory and Language , year =. doi:https://doi.org/10.1006/jmla.1999.2653 , url =

    1999

  23. [31]

    Proceedings of CogSci , year =

    Ethan Gotlieb Wilcox and Jon Gauthier and Jennifer Hu and Peng Qian and Roger Levy , title =. Proceedings of CogSci , year =

  24. [32]

    Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , year =

    Amouyal, Samuel Joseph and Meltzer-Asscher, Aya and Berant, Jonathan , title =. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , year =. doi:10.18653/v1/2025.acl-long.403 , ISBN =

    2025 doi

  25. [33]

    Cognition , year =

    Roger Levy , title =. Cognition , year =. doi:https://doi.org/10.1016/j.cognition.2007.05.006 , url =

    2007 doi

  26. [34]

    Linguistic Issues in Language Technology , year =

    Bernardy, Jean-Phillipe and Lappin, Shalom , title =. Linguistic Issues in Language Technology , year =

  27. [35]

    Gulordava, Kristina and Bojanowski, Piotr and Grave, Edouard and Linzen, Tal and Baroni, Marco , title =. Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers) , ye...

    2018

  28. [36]

    Transactions of the Association for Computational Linguistics , year =

    Linzen, Tal and Dupoux, Emmanuel and Goldberg, Yoav , title =. Transactions of the Association for Computational Linguistics , year =. doi:10.1162/tacl_a_00115 , publisher =

    doi

  29. [37]

    and Vigliocco, Gabriella , title =

    Fernandez Monsalve, Irene and Frank, Stefan L. and Vigliocco, Gabriella , title =. Proceedings of the 13th Conference of the. 2012 , editor =

    2012

  30. [38]

    and Sigman, Mariano and Lau, Ellen F

    Lago, Sol and Shalom, Diego E. and Sigman, Mariano and Lau, Ellen F. and Phillips, Colin , title =. jml , year =. doi:10.1016/j.jml.2015.02.002 , publisher =

    2015 doi

  31. [39]

    Frontiers in Psychology , year =

    Villata, Sandra and Tabor, Whitney and Franck, Julie , title =. Frontiers in Psychology , year =. doi:10.3389/fpsyg.2018.00002 , ISSN =

    2018

  32. [40]

    2012 , address =

    Tom Snijders and Roel Bosker , title =. 2012 , address =

    2012

  33. [41]

    Cognitive Science , year =

    van Schijndel, Marten and Linzen, Tal , title =. Cognitive Science , year =. doi:10.1111/cogs.12988 , publisher =

    doi

  34. [42]

    Neural language model gradients predict event-related brain potentials

    Frank, Stefan L. Neural language model gradients predict event-related brain potentials. Proceedings of the Society for Computation in Linguistics 2024. 2024

    2024

  35. [43]

    Falk Huettig and Gerry T. M. Altmann , title =. Quarterly Journal of Experimental Psychology , volume =. 2011 , doi =

    2011

  36. [44]

    Falk Huettig and Gerry T. M. Altmann , title =. Cognition , volume =. 2005 , doi =

    2005

  37. [45]

    and Yurovsky, Daniel , title =

    Zhou, Yuchen and Tarr, Michael J. and Yurovsky, Daniel , title =. PLoS ONE , volume =. 2025 , doi =

    2025

  38. [46]

    How Much Can

    Sheng Shen and Liunian Harold Li and Hao Tan and Mohit Bansal and Anna Rohrbach and Kai-Wei Chang and Zhewei Yao and Kurt Keutzer , year=. How Much Can. Proceedings of ICLR , url=

  39. [47]

    Proceedings of NeurIPS , year =

    Mariya Toneva and Leila Wehbe , title =. Proceedings of NeurIPS , year =

  40. [48]

    Advances in Neural Information Processing Systems , year =

    Brown, Tom and Mann, Benjamin and Ryder, Nick and Subbiah, Melanie and Kaplan, Jared D and Dhariwal, Prafulla and Neelakantan, Arvind and Shyam, Pranav and Sastry, Girish and Askell, Amanda and Agarwal, Sandhini and Herbert-Voss, Ariel and Krueger, Gretchen and Henighan, Tom a...

  41. [49]

    Annual Review of Linguistics , year =

    Linzen, Tal and Baroni, Marco , title =. Annual Review of Linguistics , year =

  42. [50]

    Attention is all you need , booktitle =

    Vaswani, Ashish and Shazeer, Noam and Parmar, Niki and Uszkoreit, Jakob and Jones, Llion and Gomez, Aidan N and Kaiser,. Attention is all you need , booktitle =. 2017 , pages =

    2017

  43. [51]

    Proceedings of the 40st Annual Meeting of the Cognitive Science Society , year =

    Tal Linzen and Brian Leonard , title =. Proceedings of the 40st Annual Meeting of the Cognitive Science Society , year =

  44. [52]

    Proceedings of the 42st Annual Meeting of the Cognitive Science Society , year =

    Suhas Arehalli and Tal Linzen , title =. Proceedings of the 42st Annual Meeting of the Cognitive Science Society , year =

  45. [53]

    Journal of Memory and Language , year =

    Huang, Kuan-Jung and Arehalli, Suhas and Kugemoto, Mari and Muxica, Christian and Prasad, Grusha and Dillon, Brian and Linzen, Tal , title =. Journal of Memory and Language , year =. doi:10.1016/j.jml.2024.104510 , publisher =

    2024

  46. [54]

    Proceedings of the Workshop on Cognitive Modeling and Computational Linguistics , year =

    Ryu, Soo Hyun and Lewis, Richard , title =. Proceedings of the Workshop on Cognitive Modeling and Computational Linguistics , year =

  47. [55]

    Findings of the Association for Computational Linguistics: EMNLP 2023 , year =

    Timkey, William and Linzen, Tal , title =. Findings of the Association for Computational Linguistics: EMNLP 2023 , year =

    2023

  48. [56]

    Proceedings of the 28th Conference on Computational Natural Language Learning , year =

    Bazhukov, Maxim and Voloshina, Ekaterina and Pletenev, Sergey and Anisimov, Arseny and Serikov, Oleg and Toldova, Svetlana , title =. Proceedings of the 28th Conference on Computational Natural Language Learning , year =

  49. [57]

    Cognitive Psychology , year =

    Bock, Kathryn and Miller, Carol A , title =. Cognitive Psychology , year =. doi:10.1016/0010-0285(91)90003-7 , publisher =

    doi

  50. [58]

    Semantic attraction in sentence comprehension , journal =

    Anna Laurinavichyute and. Semantic attraction in sentence comprehension , journal =. 2022 , volume =. doi:10.1111/cogs.13086 , url =

    2022 doi

  51. [59]

    Agreement attraction in grammatical sentences and the role of the task , journal =

    Laurinavichyute, Anna and. Agreement attraction in grammatical sentences and the role of the task , journal =. 2024 , volume =

    2024

  52. [60]

    Assessing the influence of attractor-verb distance on grammatical agreement in humans and language models , booktitle =

    Zacharopoulos, Christos and Desbordes, Th. Assessing the influence of attractor-verb distance on grammatical agreement in humans and language models , booktitle =. 2023 , editor =

    2023

  53. [61]

    Proceedings of the 2019 Conference of the North

    Devlin, Jacob and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina , title =. Proceedings of the 2019 Conference of the North. 2019 , pages =

    2019

  54. [62]

    Proceedings of NeurIPS , year =

    Maximilian Beck and Korbinian Pöppel and Markus Spanring and Andreas Auer and Oleksandra Prudnikova and Michael K Kopp and Günter Klambauer and Johannes Brandstetter and Sepp Hochreiter , title =. Proceedings of NeurIPS , year =

  55. [63]

    Society for Computation in Linguistics , year =

    Kim, Sanghee and Davis, Forrest , title =. Society for Computation in Linguistics , year =. doi:10.7275/SCIL.3156 , copyright =

    doi

  56. [64]

    and Levy, Roger , title =

    Smith, Nathaniel J. and Levy, Roger , title =. Cognition , year =

  57. [65]

    Proceedings of the National Academy of Sciences , year =

    Shain, Cory and Meister, Clara and Pimentel, Tiago and Cotterell, Ryan and Levy, Roger , title =. Proceedings of the National Academy of Sciences , year =. doi:10.1073/pnas.2307876121 , publisher =

    doi

  58. [66]

    Boston and John T

    Marisa F. Boston and John T. Hale and Reinhold Kliegl and Umesh Patil and Shravan Vasishth , title =. Journal of Eye Movement Research , year =

  59. [67]

    Journal of Memory and Language , year =

    Harald Baayen and Shravan Vasishth and Reinhold Kliegl and Douglas Bates , title =. Journal of Memory and Language , year =. doi:https://doi.org/10.1016/j.jml.2016.11.006 , url =

    2016 doi

  60. [68]

    Proceedings of ACL , year =

    Ryo Yoshida and Shinnosuke Isono and Taiga Someya and Yohei Oseki and Tatsuki Kuribayashi , title =. Proceedings of ACL , year =

  61. [69]

    Cognition , year =

    Vera Demberg and Frank Keller , title =. Cognition , year =

  62. [70]

    Language and Cognitive Processes , year = 2002, volume = 17, number = 4, pages =

    Franck, Julie and Vigliocco, Gabriella and Nicol, Janet , title =. Language and Cognitive Processes , year = 2002, volume = 17, number = 4, pages =. doi:10.1080/01690960143000254 , publisher =

    2002 doi

  63. [71]

    Journal of Memory and Language , year =

    Staub, Adrian , title =. Journal of Memory and Language , year =. doi:10.1016/j.jml.2008.11.002 , publisher =

    2008 doi

  64. [72]

    Language and Linguistics Compass , year =

    Portelance, Eva and Jasbi, Masoud , title =. Language and Linguistics Compass , year =. doi:https://doi.org/10.1111/lnc3.70001 , url =

    doi

  65. [73]

    Proceedings of the Annual Meeting of the Cognitive Science Society , year =

    Degen, Judith and Kursat, Leyla and Leigh, Daisy Dorothy , title =. Proceedings of the Annual Meeting of the Cognitive Science Society , year =

  66. [74]

    6", number =

    Cooper, Roger M , title =. Cognitive psychology , volume = "6", number = "1", pages = "84-107", year =

  67. [75]

    Discourse Processes , year =

    Yoon, Si On and Brown-Schmidt, Sarah , title =. Discourse Processes , year =

  68. [76]

    Machine Learning for Healthcare Conference , year =

    Zhang, Yuhao and Jiang, Hang and Miura, Yasuhide and Manning, Christopher D and Langlotz, Curtis P , title =. Machine Learning for Healthcare Conference , year =

  69. [77]

    and Horowitz, Todd S

    Wolfe, Jeremy M. and Horowitz, Todd S. , title =. Nature Human Behaviour , year =

  70. [78]

    James Murdoch and Chandan Singh and Karl Kumbier and Reza Abbasi-Asl and Bin Yu , title =

    W. James Murdoch and Chandan Singh and Karl Kumbier and Reza Abbasi-Asl and Bin Yu , title =. Proceedings of the National Academy of Sciences , year =

  71. [79]

    2024 , isbn =

    Hindennach, Susanne and Shi, Lei and Bulling, Andreas , title =. 2024 , isbn =. doi:10.1145/3649902.3656356 , booktitle =

    2024

  72. [80]

    Barr and Roger Levy and Christoph Scheepers and Harry J

    Dale J. Barr and Roger Levy and Christoph Scheepers and Harry J. Tily , keywords =. Random effects structure for confirmatory hypothesis testing: Keep it maximal , journal =. 2013 , issn =. doi:https://doi.org/10.1016/j.jml.2012.11.001 , url =

    2013 doi

  73. [81]

    and McClelland, James L

    Rumelhart, David E. and McClelland, James L. , title =. Parallel Distributed Processing: Explorations in the Microstructure of Cognition, Volume 2: Psychological and Biological Models , editor =

  74. [82]

    and Magnuson, James S

    Allopenna, Paul D. and Magnuson, James S. and Tanenhaus, Michael K. , title =. Journal of Memory and Language , volume =. 1998 , doi =

    1998

  75. [83]

    and Elman, Jeffrey L

    McClelland, James L. and Elman, Jeffrey L. , title =. Cognitive Psychology , volume =. 1986 , doi =

    1986

  76. [84]

    Sequence Classification with Human Attention

    Barrett, Maria and Bingel, Joachim and Hollenstein, Nora and Rei, Marek and S gaard, Anders. Sequence Classification with Human Attention. Proceedings of the 22nd Conference on Computational Natural Language Learning. 2018. doi:10.18653/v1/K18-1030

    2018 doi

  77. [85]

    Controlling Reading Ease with Gaze-Guided Text Generation

    S. Controlling Reading Ease with Gaze-Guided Text Generation. Proceedings of the 19th Conference of the E uropean Chapter of the A ssociation for C omputational L inguistics (Volume 1: Long Papers). 2026. doi:10.18653/v1/2026.eacl-long.107

    2026 doi

  78. [86]

    Proceedings of the 31st International Conference on Computational Linguistics , year =

    Tilli, Pascal and Vu, Ngoc Thang , title =. Proceedings of the 31st International Conference on Computational Linguistics , year =

  79. [87]

    Proceedings of NeurIPS , year =

    Liu, Haotian and Li, Chunyuan and Wu, Qingyang and Lee, Yong Jae , title =. Proceedings of NeurIPS , year =

  80. [88]

    arXiv preprint arXiv:2502.13923 , year =

    Bai, Shuai and Chen, Keqin and Liu, Xuejing and Wang, Jialin and Ge, Wenbin and Song, Sibo and Dang, Kai and Wang, Peng and Wang, Shijie and Tang, Jun and Zhong, Humen and Zhu, Yuanzhi and Yang, Mingkun and Li, Zhaohai and Wan, Jianqiang and Wang, Pengfei and Ding, Wei and Fu,...

    arXiv

  81. [89]

    Proceedings of ICLR , year=

    Data Filtering Networks , author=. Proceedings of ICLR , year=

  82. [90]

    Proceedings of NeurIPS -- Datasets and Benchmarks Track , year=

    DataComp: In search of the next generation of multimodal datasets , author=. Proceedings of NeurIPS -- Datasets and Benchmarks Track , year=. 2304.14108 , archivePrefix=

    arXiv

  83. [91]

    Proceedings of the IEEE international conference on computer vision , year =

    Antol, Stanislaw and Agrawal, Aishwarya and Lu, Jiasen and Mitchell, Margaret and Batra, Dhruv and Zitnick, C Lawrence and Parikh, Devi , title =. Proceedings of the IEEE international conference on computer vision , year =

  84. [92]

    Proceedings of the 38th International Conference on Machine Learning , year =

    Radford, Alec and Kim, Jong Wook and Hallacy, Chris and Ramesh, Aditya and Goh, Gabriel and Agarwal, Sandhini and Sastry, Girish and Askell, Amanda and Mishkin, Pamela and Clark, Jack and Krueger, Gretchen and Sutskever, Ilya , title =. Proceedings of the 38th International Co...

  85. [93]

    , title =

    Magnuson, James S. , title =. Journal of Cultural Cognitive Science , year =. doi:10.1007/s41809-019-00035-3 , url =

    doi

  86. [94]

    Behavior Research Methods , year =

    Ito, Aine and Knoeferle, Pia , title =. Behavior Research Methods , year =. doi:10.3758/s13428-022-01969-3 , url =

    doi

  87. [95]

    Cooper and Eberhard, Kathleen M

    Bock, Kathryn and Butterfield, Sally and Cutler, Anne and Cutting, J. Cooper and Eberhard, Kathleen M. and Humphreys, Karin R. , title =. Language , year =. doi:10.1353/lan.2006.0011 , publisher =

    2006

  88. [96]

    Proceedings of the Fifth Annual Texas Linguistics Society Conference , year =

    Pfau, Roland , title =. Proceedings of the Fifth Annual Texas Linguistics Society Conference , year =

  89. [97]

    Acta psychologica , year =

    Huettig, Falk and Rommers, Joost and Meyer, Antje S , title =. Acta psychologica , year =

  90. [98]

    and Magnuson, James S

    Mirman, Daniel and Dixon, James A. and Magnuson, James S. , title =. Journal of Memory and Language , volume =. 2008 , month =

    2008

  91. [99]

    2015 , issn =

    On computational modeling of visual saliency: Examining what’s right, and what’s left , journal =. 2015 , issn =. doi:https://doi.org/10.1016/j.visres.2015.01.010 , url =

    2015 doi

  92. [100]

    Science , url =

    Tanenhaus, Michael K and Spivey-Knowlton, Michael J and Eberhard, Kathleen M and Sedivy, Julie C , title =. Science , url =. 1995 , volume =

    1995

  93. [101]

    Cognition , year =

    Sedivy, Julie C and Tanenhaus, Michael K and Chambers, Craig G and Carlson, Greg N , title =. Cognition , year =

  94. [102]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year =

    Cherti, Mehdi and Beaumont, Romain and Wightman, Ross and Wortsman, Mitchell and Ilharco, Gabriel and Gordon, Cade and Schuhmann, Christoph and Schmidt, Ludwig and Jitsev, Jenia , title =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)...

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.