# GPT2MEG: Quantizing MEG for Autoregressive Generation

arXiv: https://arxiv.org/abs/2404.09256 · HTML: https://mlsanity.com/paper/2404.09256/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 28 references that are in the map.
- First posted on arXiv 14 Apr 2024.

## Builds on (most cited first, 12 of 12)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [WaveNet: A Generative Model for Raw Audio](https://mlsanity.com/paper/1609.03499/) (2016)
- [Chronos: Learning the Language of Time Series](https://mlsanity.com/paper/2403.07815/) (2024)
- [A decoder-only foundation model for time-series forecasting](https://mlsanity.com/paper/2310.10688/) (2023)
- [BENDR: using transformers and a contrastive self-supervised learning task to learn from massive amounts of EEG data](https://mlsanity.com/paper/2101.12037/) (2021)
- [Uncovering the structure of clinical EEG signals with self-supervised learning](https://mlsanity.com/paper/2007.16104/) (2020)
- [Neuro-GPT: Towards A Foundation Model for EEG](https://mlsanity.com/paper/2311.03764/) (2023)
- [BrainBERT: Self-supervised representation learning for intracranial recordings](https://mlsanity.com/paper/2302.14367/) (2023)
- [Group-level Brain Decoding with Deep Learning](https://mlsanity.com/paper/2205.14102/) (2022)
- [Deep Recurrent Encoder: A scalable end-to-end network to model brain signals](https://mlsanity.com/paper/2103.02339/) (2021)

## Built on by (most cited first, 3 of 3)

- [Let EEG Models Learn EEG](https://mlsanity.com/paper/2605.21280/) (2026)
- [OmniMouse: Scaling properties of multi-modal, multi-task Brain Models on 150B Neural Tokens](https://mlsanity.com/paper/2604.18827/) (2026)
- [A Roadmap for MEG Foundation Models](https://mlsanity.com/paper/2609.04461/) (2026)

## Abstract (arXiv)

Foundation models trained with self-supervised objectives are increasingly applied to brain recordings, but autoregressive generation of realistic multichannel neural time series remains comparatively underexplored, particularly for Magnetoencephalography (MEG). We study (i) modified multichannel WaveNet variants and (ii) a GPT-2-style Transformer, autoregressively trained by next-step prediction on unlabelled MEG. For the Transformer, we propose a simple quantization/tokenization and embedding scheme (channel, subject, and task-condition embeddings) that repurposes a language-model architecture for continuous, high-rate multichannel time series and enables conditional simulation of task-evoked activity. Across forecasting, long-horizon generation, and downstream decoding, GPT2MEG more faithfully reproduces temporal, spectral, and task-evoked statistics of real MEG than WaveNet variants and linear autoregressive baselines, and scales to multiple subjects via subject embeddings. Code available at https://github.com/ricsinaruto/MEG-transfer-decoding.