∆ DELTA: Language Diffusion-based EEG-to-Text Architecture

mingyu jeon, Hyobin Kim

Keywords: EEG-to-Text, Residual Vector Quantization, Language Diffusion Model, Discrete Tokenization, Multimodal Brain-Language Learning

TL;DR: We propose DELTA, an EEG-to-text architecture that converts continuous brain signals into multi-layer discrete tokens and uses a non-sequential diffusion model to reconstruct text, overcoming the cumulative errors of traditional sequential methods.

Abstract:

Electroencephalogram (EEG)-to-text remains challenging due to high-dimensional
noise, subject variability, and error accumulation in autoregressive decoding. We in-
troduce DELTA, which pairs a Residual Vector Quantization (RVQ) EEG tokenizer
with a masked language diffusion model (LLaDA). RVQ discretizes continuous
EEG into multi-layer tokens to reduce noise and individual differences, while
LLaDA reconstructs sentences via non-sequential denoising. On ZuCo, DELTA
improves semantic alignment by up to 5.37 points over autoregressive baselines,
achieving BLEU-1 21.9 and ROUGE-1 F 17.2 under word-level conditions. These
results enable reliable text generation from small EEG-text datasets and point
toward scalable multimodal EEG-language models.

Submission Number: 42