mlsanityData to 14 Oct 2026

Training language models to summarize narratives improves brain alignment

arXiv 2212.10898 ↗First posted 21 Dec 2022cs.CL q-bio.NCcited by 26

Not on the map. Why?

Abstract (arXiv)

Building systems that achieve a deeper understanding of language is one of the central goals of natural language processing (NLP). Towards this goal, recent works have begun to train language models on narrative datasets which require extracting the most critical information by integrating across long contexts. However, it is still an open question whether these models are learning a deeper understanding of the text, or if the models are simply learning a heuristic to complete the task. This work investigates this further by turning to the one language processing system that truly understands complex language: the human brain. We show that training language models for deeper narrative understanding results in richer representations that have improved alignment to human brain activity. We further find that the improvements in brain alignment are larger for character names than for other discourse features, which indicates that these models are learning important narrative elements. Taken together, these results suggest that this type of training can indeed lead to deeper language understanding. These findings have consequences both for cognitive neuroscience by revealing some of the significant factors behind brain-NLP alignment, and for NLP by highlighting that understanding of long-range context can be improved beyond language modeling.

Key facts. Cited by 26 papers in the map (data through 14 Oct 2026). Builds on 18 of its 37 references that are in the map. First posted on arXiv 21 Dec 2022.

Builds on

18 of 37 references.

References · 18

  1. 2019BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehensioncited by 5,893
  2. 2020Longformer: The Long-Document Transformercited by 3,413
  3. 2020Big Bird: Transformers for Longer Sequencescited by 1,684
  4. 2019Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inferencecited by 876
  5. 2017The NarrativeQA Reading Comprehension Challengecited by 597
  6. 2021LongT5: Efficient Text-To-Text Transformer for Long Sequencescited by 232
  7. 2021BookSum: A Collection of Datasets for Long-form Narrative Summarizationcited by 184
  8. 2019Interpreting and improving natural-language processing (in machines) with natural language-processing (in the brain)cited by 128
  9. 2020How Can We Accelerate Progress Towards Human-like Linguistic Generalization?cited by 114
  10. 2020Syntactic Data Augmentation Increases Robustness to Inference Heuristicscited by 86
  11. 2020On the Interplay Between Fine-tuning and Sentence-level Probing for Linguistic Knowledge in Pre-trained Transformerscited by 39
  12. 2019Inducing brain-relevant bias in natural language processing modelscited by 38
6 more
  1. 2022Neural Language Taskonomy: Which NLP Tasks are the most Predictive of fMRI Brain Activity?cited by 27
  2. 2022Neural Language Models are not Born Equal to Fit Brain Data, but Training Helpscited by 26
  3. 2022Language models and brains align due to more than next-word prediction and word-level informationcited by 14
  4. 2021Does injecting linguistic structure into language models lead to better alignment with brain recordings?cited by 12
  5. 2022Same Cause; Different Effects in the Braincited by 10
  6. 2022A Survey of Machine Narrative Reading Comprehension Assessmentscited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 8; 2024: 6; 2025: 10; 2026: 2.

26 papers.

Sort:

By area:

  1. 2023Getting aligned on representational alignmentcited by 143
  2. 2023Scaling laws for language encoding models in fMRIcited by 68
  3. 2024Large Language Models and Cognitive Science: A Comprehensive Review of Similarities, Differences, and Challengescited by 49
  4. 2023Instruction-tuning Aligns LLMs to the Human Braincited by 40
  5. 2023Deep Neural Networks and Brain Alignment: Brain Encoding and Decoding (Survey)cited by 21
  6. 2024Improving Semantic Understanding in Speech Language Models via Brain-tuningcited by 20
  7. 2023Speech language models lack important brain-relevant semanticscited by 19
  8. 2024Large Language Model Benchmarks in Medical Taskscited by 17
  9. 2024What Are Large Language Models Mapping to in the Brain? A Case Against Over-Reliance on Brain Scorescited by 10
  10. 2023Mapping Brains with Language Models: A Surveycited by 8
  11. 2025Multi-modal brain encoding models for multi-modal stimulicited by 8
  12. 2023BrainLLM: Generative Language Decoding from Brain Recordingscited by 7
14 more
  1. 2025Brain-Language Model Alignment: Insights into the Platonic Hypothesis and Intermediate-Layer Advantagecited by 5
  2. 2024Computational Models to Study Language Processing in the Human Brain: A Surveycited by 4
  3. 2025Explanations of Large Language Models Explain Language Representations in the Braincited by 4
  4. 2025Triple Phase Transitions: Understanding the Learning Dynamics of Large Language Models from a Neuroscience Perspectivecited by 4
  5. 2025Correlating instruction-tuning (in multimodal models) with vision-language processing (in the brain)cited by 4
  6. 2025Language models align with brain regions that represent concepts across modalitiescited by 4
  7. 2025Far from the Shallow: Brain-Predictive Reasoning Embedding through Residual Disentanglementcited by 4
  8. 2023Divergences between Language Models and Human Brainscited by 3
  9. 2025Fine-grained Analysis of Brain-LLM Alignment through Input Attributioncited by 3
  10. 2024Inducing Human-like Biases in Moral Reasoning Language Modelscited by 2
  11. 2025CLIP is All You Need for Human-like Semantic Representations in Stable Diffusioncited by 0
  12. 2025Better audio representations are more brain-like: linking model-brain alignment with performance in downstream auditory taskscited by 0
  13. 2026Linguistic properties and model scale in brain encoding: from small to compressed language modelscited by 0
  14. 2026How does longer temporal context enhance multimodal narrative video processing in the brain?cited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Do Large Language Models Think Like the Brain? Sentence-Level Evidences from Layer-Wise Embeddings and fMRI
  2. 2026Cognitive Expert Language Models Better Align with the Corresponding Brain Systems
  3. 2026When Language Models Lose Their Mind: The Consequences of Brain Misalignment
  4. 2026Beyond representational alignment with brain-guided language models for robust reasoning
  5. 2026Interpreting Brain Responses to Language with Sparse Features from Language Models
  6. 2026Do VLMs Align Better with Humans than LLMs during Natural Reading?
  7. 2024Vision-Language Models Align with Human Neural Representations in Concept Processing
  8. 2025Improve Language Model and Brain Alignment via Associative Memory