mlsanityData to 7 Oct 2026

A Dual-Task Paradigm to Investigate Sentence Comprehension Strategies in Language Models

arXiv 2604.26351 ↗First posted 29 Apr 2026cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

Language models (LMs) behave more like humans when their cognitive resources are restricted, particularly in predicting sentence processing costs such as reading times. However, it remains unclear whether such constraints similarly affect sentence comprehension strategies. Besides, existing methods do not directly target the balance between memory storage and sentence processing, which is central to human working memory. To address this issue, we propose a dual-task paradigm that combines an arithmetic computation task with a sentence comprehension task, such as "The 2 cocktail + blended 3 =..." Our experiments show that under dual-task conditions, GPT-4o, o3-mini, and o4-mini shift toward plausibility-based comprehension, mirroring humans' rational inference. Specifically, these models show a greater accuracy gap between plausible sentences (e.g., "The cocktail was blended by the bartender") and implausible sentences (e.g., "The bartender was blended by the cocktail") in the dual-task condition compared to the single-task conditions. These findings suggest that constraints on the balance between memory and processing resources promote rational inference in LMs. More broadly, they support the view that human-like sentence comprehension fundamentally arises from the allocation of limited cognitive resources.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 19 of its 61 references that are in the map. First posted on arXiv 29 Apr 2026.

Builds on

19 of 61 references.

References · 19

  1. 2022Large Language Models are Zero-Shot Reasonerscited by 4,323
  2. 2020Shortcut Learning in Deep Neural Networkscited by 1,928
  3. 2023Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formattingcited by 778
  4. 2024OLMo: Accelerating the Science of Language Modelscited by 599
  5. 2017Adversarial Examples for Evaluating Reading Comprehension Systemscited by 568
  6. 2022Why Does Surprisal From Larger Transformer-Based Language Models Provide a Poorer Fit to Human Reading Times?cited by 111
  7. 2023Batch Prompting: Efficient Inference with Large Language Model APIscited by 84
  8. 2020Look at the First Sentence: Position Bias in Question Answeringcited by 83
  9. 2023Cognitive Overload: Jailbreaking Large Language Models with Overloaded Logical Thinkingcited by 78
  10. 2023Large Language Models Can be Lazy Learners: Analyze Shortcuts in In-Context Learningcited by 69
  11. 2021Lower Perplexity is Not Always Human-Likecited by 52
  12. 2022Context Limitations Make Neural Language Models More Human-Likecited by 32
7 more
  1. 2024Multi-Task Inference: Can Large Language Models Follow Multiple Instructions at Once?cited by 28
  2. 2021Accounting for Agreement Phenomena in Sentence Comprehension with Transformer Language Models: Effects of Similarity-based Interference on Surprisal and Attentioncited by 26
  3. 2023Working Memory Capacity of ChatGPT: An Empirical Studycited by 23
  4. 2023A Language Model with Limited Memory Capacity Captures Interference in Human Sentence Processingcited by 22
  5. 2025If Attention Serves as a Cognitive Model of Human Memory Retrieval, What is the Plausible Memory Representation?cited by 4
  6. 2024What Makes Language Models Good-enough?cited by 2
  7. 2025Comparing Human and Language Models Sentence Processing Difficulties on Complex Structurescited by 2

Built on by

Not cited yet.