The study examines how reader proficiency influences the relationship between layer-wise surprisal from large language models (LLMs) and eye-tracking gaze measures. Using the MECO L2 corpus, researchers compared high- and low-proficiency readers on first-pass gaze duration (FPGD) and total gaze duration (TGD), finding that lower-proficiency readers exhibit deeper Predictive Depth for FPGD, while TGD shows deeper Predictive Depth across both groups. The results suggest that the distribution of predictive power across LLM layers relates to the timing and breadth of reading processes and varies with reader proficiency.
By Akio Hayakawa, Horacio Saggion
arXiv:2606.27206v2 Announce Type: replace
Abstract: Garden path sentences present a processing difficulty for humans--- the sentence prefix leads the listener towards one interpretation, until the li...
By Alan Zhou, Milo\v{s} Stanojevi\'c, John T. Hale
arXiv:2607. 08152v1 Announce Type: cross Abstract: On the recent EyeBench benchmark, predicting reading comprehension from eye movements exposes a stark gap: text-aware models using pretrained language models reach 56--63% AUROC, while gaze-only models operate at chance.
By Sumin Lee, Kyeonghun Kim, Subeen Lee, Jiwon Yang, Tien Nguyen, Ken Ying-Kai Liao, Nam-Joon Kim
arXiv:2603.04419v3 Announce Type: replace-cross
Abstract: Vision-language models produce different object and use descriptions under different persona prompts, but low overlap alone does not identify...
By Murad Farzulla
The paper investigates energy-based transformers as predictors of reading difficulty, extending the use of transformer language models in psycholinguistics. It demonstrates that the energy measure from these models robustly predicts reading times across multiple corpora, outperforming traditional metrics like surprisal and attention entropy. In a controlled experiment on relative clause processing, energy captures known asymmetries, suggesting it may unify previously complementary predictors.
By Jakub Dotlacil, Ece Takmaz
arXiv:2501. 06286v2 Announce Type: replace-cross Abstract: Multi-hop question answering requires a system to identify and integrate evidence distributed across documents, yet large language models remain vulnerable to irrelevant context.
By Iman Barati, Arash Ghafouri, Behrouz Minaei-Bidgoli
arXiv:2608. 15022v1 Announce Type: new Abstract: Language models hold latent quantities in a form they can report on, and more of a quantity is present in that form when the task requires reusing it flexibly.
By Parsa Mazaheri
Multimodal LLMs can see a document, but they often can't read it reliably. Small text, tables, visual cues, and topological elements still trip them up under direct visual inference, even when the pag...
Surprisal theory holds that the human processing difficulty of a linguistic unit in context is an affine function of its surprisal under some language model. I argue this claim is a tautology without further constraint: for any non-negative difficulty measure over units in context, there exists a language model whose surprisal is an affine function of it under mild technical conditions.
arXiv:2608. 19739v1 Announce Type: cross Abstract: Multimodal LLMs can see a document, but they often can't read it reliably.
By Alin-Ionut Popa
arXiv:2608. 11138v1 Announce Type: cross Abstract: We propose that a model's uncertainty about a token is reflected not only in the breadth of its output distribution but also in whether a confident prediction is \emph{fragile} under perturbation of its attention pathways.
By Minsoo Kim, Sungyoung Ji, Kisung Moon, Ilyong Yoon
The study compares two ways of obtaining predictions from language models fine‑tuned on customer behavior: scoring answer tokens directly versus generating a written rationale and then scoring the resulting answer. Across 13 model‑domain cells covering four retail tasks, scored readouts consistently rank outcomes more accurately than generated readouts, with an AUC improvement ranging from 1.5 to 14.5 points. The authors also find that a third readout—eliciting a probability before any verdict—improves calibration but only when outcome rates are represented in training, and they recommend using generated rationales for interpretability while relying on scored heads for ranking.
By Touchapon Kraisingkorn, Krittin Pachtrachai, Wachiravit Modecrua