mlsanityData to 7 Oct 2026

Fine-grained Analysis of Brain-LLM Alignment through Input Attribution

arXiv 2510.12355 ↗First posted 14 Oct 2025cs.CLcited by 3

On the map: LLM alignment / RLHF / preference optimisation → LLM preference optimization methods

Abstract (arXiv)

Understanding the alignment between large language models (LLMs) and human brain activity can reveal computational principles underlying language processing. We introduce a fine-grained input attribution method to identify the specific words most important for brain-LLM alignment, and leverage it to study a contentious research question about brain-LLM alignment: the relationship between brain alignment (BA) and next-word prediction (NWP). Our findings reveal that BA and NWP rely on largely distinct word subsets: NWP exhibits recency and primacy biases with a focus on syntax, while BA prioritizes semantic and discourse-level information with a more targeted recency effect. This work advances our understanding of how LLMs relate to human language processing and highlights differences in feature reliance between BA and NWP. Beyond this study, our attribution method can be broadly applied to explore the cognitive relevance of model predictions in diverse language processing tasks.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Placed in LLM preference optimization methods, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 23 of its 41 references that are in the map. First posted on arXiv 14 Oct 2025.

Builds on

23 of 41 references.

References · 23

  1. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,616
  2. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,581
  3. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,383
  4. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  5. 2021Efficiently Modeling Long Sequences with Structured State Spacescited by 2,424
  6. 2019HuggingFace's Transformers: State-of-the-art Natural Language Processingcited by 2,201
  7. 2020Captum: A unified and generic model interpretability library for PyTorchcited by 542
  8. 2016Not Just a Black Box: Learning Important Features Through Propagating Activation Differencescited by 295
  9. 2019Interpreting and improving natural-language processing (in machines) with natural language-processing (in the brain)cited by 128
  10. 2025On the Emergence of Position Bias in Transformerscited by 57
  11. 2023Instructed to Bias: Instruction-Tuned Language Models Exhibit Emergent Cognitive Biascited by 51
  12. 2022Joint processing of linguistic properties in brains and language modelscited by 37
11 more
  1. 2024Contextual Feature Extraction Hierarchies Converge in Large Language Models and the Braincited by 32
  2. 2025From Language to Cognition: How LLMs Outgrow the Human Language Networkcited by 30
  3. 2022Training language models to summarize narratives improves brain alignmentcited by 26
  4. 2024Improving Semantic Understanding in Speech Language Models via Brain-tuningcited by 20
  5. 2023Speech language models lack important brain-relevant semanticscited by 19
  6. 2024Understanding and Mitigating Bottlenecks of State Space Models through the Lens of Recency and Over-smoothingcited by 17
  7. 2022Language models and brains align due to more than next-word prediction and word-level informationcited by 14
  8. 2023Token-wise Decomposition of Autoregressive Language Model Hidden States for Analyzing Model Predictionscited by 6
  9. 2025Exploiting Primacy Effect To Improve Large Language Modelscited by 5
  10. 2025Emergence of the Primacy Effect in Structured State-Space Modelscited by 4
  11. 2025Explanations of Large Language Models Explain Language Representations in the Braincited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026Linguistics and Human Brain: A Perspective of Computational Neurosciencecited by 3
  2. 2026Layer-wise Positional Bias in Short-Context Language Modelingcited by 1
  3. 2026Training-Driven Representational Geometry Modularization Predicts Brain Alignment in Language Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026When Language Models Lose Their Mind: The Consequences of Brain Misalignment
  2. 2023Instruction-tuning Aligns LLMs to the Human Brain
  3. 2026Linguistic properties and model scale in brain encoding: from small to compressed language models
  4. 2026Brain-LLM Alignment Tracks Training Data, Not Typology
  5. 2026Cognitive Expert Language Models Better Align with the Corresponding Brain Systems
  6. 2026Beyond representational alignment with brain-guided language models for robust reasoning
  7. 2025Do Large Language Models Think Like the Brain? Sentence-Level Evidences from Layer-Wise Embeddings and fMRI
  8. 2026Do Language Models Align with Brains? Prediction Scores Are Not Enough