@inproceedings{mishra-etal-2025-thought2text,
title = "{T}hought2{T}ext: Text Generation from {EEG} Signal using Large Language Models ({LLM}s)",
author = "Mishra, Abhijit and
Shukla, Shreya and
Torres, Jose and
Gwizdka, Jacek and
Roychowdhury, Shounak",
editor = "Chiruzzo, Luis and
Ritter, Alan and
Wang, Lu",
booktitle = "Findings of the Association for Computational Linguistics: NAACL 2025",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-naacl.207/",
doi = "10.18653/v1/2025.findings-naacl.207",
pages = "3747--3759",
ISBN = "979-8-89176-195-7",
abstract = "Decoding and expressing brain activity in a comprehensible form is a challenging frontier in AI. This paper presents *Thought2Text*, which uses instruction-tuned Large Language Models (LLMs) fine-tuned with EEG data to achieve this goal. The approach involves three stages: (1) training an EEG encoder for visual feature extraction, (2) fine-tuning LLMs on image and text data, enabling multimodal description generation, and (3) further fine-tuning on EEG embeddings to generate text directly from EEG during inference. Experiments on a public EEG dataset collected for six subjects with image stimuli and text captions demonstrate the efficacy of multimodal LLMs (*LLaMA-v3*, *Mistral-v0.3*, *Qwen2.5*), validated using traditional language generation evaluation metrics, as well as *fluency* and *adequacy* measures. This approach marks a significant advancement towards portable, low-cost ``thoughts-to-text'' technology with potential applications in both neuroscience and natural language processing."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="mishra-etal-2025-thought2text">
<titleInfo>
<title>Thought2Text: Text Generation from EEG Signal using Large Language Models (LLMs)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Abhijit</namePart>
<namePart type="family">Mishra</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shreya</namePart>
<namePart type="family">Shukla</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jose</namePart>
<namePart type="family">Torres</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jacek</namePart>
<namePart type="family">Gwizdka</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shounak</namePart>
<namePart type="family">Roychowdhury</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: NAACL 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Luis</namePart>
<namePart type="family">Chiruzzo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alan</namePart>
<namePart type="family">Ritter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Albuquerque, New Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-195-7</identifier>
</relatedItem>
<abstract>Decoding and expressing brain activity in a comprehensible form is a challenging frontier in AI. This paper presents *Thought2Text*, which uses instruction-tuned Large Language Models (LLMs) fine-tuned with EEG data to achieve this goal. The approach involves three stages: (1) training an EEG encoder for visual feature extraction, (2) fine-tuning LLMs on image and text data, enabling multimodal description generation, and (3) further fine-tuning on EEG embeddings to generate text directly from EEG during inference. Experiments on a public EEG dataset collected for six subjects with image stimuli and text captions demonstrate the efficacy of multimodal LLMs (*LLaMA-v3*, *Mistral-v0.3*, *Qwen2.5*), validated using traditional language generation evaluation metrics, as well as *fluency* and *adequacy* measures. This approach marks a significant advancement towards portable, low-cost “thoughts-to-text” technology with potential applications in both neuroscience and natural language processing.</abstract>
<identifier type="citekey">mishra-etal-2025-thought2text</identifier>
<identifier type="doi">10.18653/v1/2025.findings-naacl.207</identifier>
<location>
<url>https://aclanthology.org/2025.findings-naacl.207/</url>
</location>
<part>
<date>2025-04</date>
<extent unit="page">
<start>3747</start>
<end>3759</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Thought2Text: Text Generation from EEG Signal using Large Language Models (LLMs)
%A Mishra, Abhijit
%A Shukla, Shreya
%A Torres, Jose
%A Gwizdka, Jacek
%A Roychowdhury, Shounak
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Findings of the Association for Computational Linguistics: NAACL 2025
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-195-7
%F mishra-etal-2025-thought2text
%X Decoding and expressing brain activity in a comprehensible form is a challenging frontier in AI. This paper presents *Thought2Text*, which uses instruction-tuned Large Language Models (LLMs) fine-tuned with EEG data to achieve this goal. The approach involves three stages: (1) training an EEG encoder for visual feature extraction, (2) fine-tuning LLMs on image and text data, enabling multimodal description generation, and (3) further fine-tuning on EEG embeddings to generate text directly from EEG during inference. Experiments on a public EEG dataset collected for six subjects with image stimuli and text captions demonstrate the efficacy of multimodal LLMs (*LLaMA-v3*, *Mistral-v0.3*, *Qwen2.5*), validated using traditional language generation evaluation metrics, as well as *fluency* and *adequacy* measures. This approach marks a significant advancement towards portable, low-cost “thoughts-to-text” technology with potential applications in both neuroscience and natural language processing.
%R 10.18653/v1/2025.findings-naacl.207
%U https://aclanthology.org/2025.findings-naacl.207/
%U https://doi.org/10.18653/v1/2025.findings-naacl.207
%P 3747-3759
Markdown (Informal)
[Thought2Text: Text Generation from EEG Signal using Large Language Models (LLMs)](https://aclanthology.org/2025.findings-naacl.207/) (Mishra et al., Findings 2025)
ACL