Encoding and Decoding Language in the Brain with Language Models
Paper · 2026 · Anuja Negi, Mathis Lamarre, Christine Tseng et al.
Summary
This is a two-page abstract for a half-day tutorial at EACL 2026, not an empirical study. Its subject is NeuroAI, the field that uses language models to explain and predict human brain activity during language comprehension. The tutorial begins with the basics of brain–model alignment. That covers encoding models, which predict neural responses from model representations, and methods for comparing brain activity with those representations. It then surveys newer directions: how scaling laws of language models relate to brain alignment, encoding across several languages, fine-tuning models with brain recordings, and decoding. Decoding here includes reconstructing the meaning of continuous language from brain signals. Attendees are promised an overview of naturalistic datasets and computational frameworks used in the area. The abstract lists five learning objectives but reports no new data, experiments, or quantitative results, and it names no specific corpora or models. For the library, its value is as a compact map of current research threads in brain–LM alignment and as a pointer to the TU Berlin group presenting it.
Key points
- The tutorial is listed as T3 in the EACL 2026 Proceedings, Volume 6: Tutorial Abstracts, pp. 7–8. It is labelled introductory and runs Sunday, March 29, 2026, 09:00–12:30, in Pavillon DE RABAT (Level 1).
- The presenters are Anuja Negi (written "Negim" in the PDF extraction), Mathis Lamarre, Christine Tseng and Subba Reddy Oota. All four work at TU Berlin in a lab named for cognitive computing across both biological and artificial systems, and Negi and Lamarre are also affiliated with BCCN Berlin.
- Topics covered:
- foundations of brain–LM alignment
- scaling laws for brain alignment
- multilingual brain encoding
- brain-informed fine-tuning
- semantic decoding of continuous language
- The five learning objectives are:
- alignment and encoding concepts
- methods for comparing brain activity with model representations
- multilingual processing in brains and models
- brain-informed fine-tuning and evaluation
- applications and future directions
- Tutorial materials are at subbareddy248.github.io/tutorials/eacl-2026-brain-ai/. No datasets, models or metrics are named in the abstract itself.
Fields, topics and concepts
📚 Shelves 6The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksNeuroscience · 165 worksCognitive Neuroscience · 164 works
🧭 Topics 1Research subjects and languages the work is about.Language models & pretraining · 3,320 works
Cite
Cite as Anuja Negi, Mathis Lamarre, Christine Tseng and Subba Reddy Oota. 2026. Encoding and Decoding Language in the Brain with Language Models. Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 6: Tutorial Abstracts).
@inproceedings{negi2026encoding,
title = {{Encoding and Decoding Language in the Brain with Language Models}},
author = {Anuja Negi, Mathis Lamarre, Christine Tseng and Subba Reddy Oota},
booktitle = {Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 6: Tutorial Abstracts)},
year = {2026},
doi = {10.18653/v1/2026.eacl-tutorials.3},
url = {https://doi.org/10.18653/v1/2026.eacl-tutorials.3}
}Updated
This page in Markdown → https://md.research.elamur.ai/free/en/papers/negi-2026-encoding-decoding-language-brain-language/