# Temporal Context and Architecture: A Benchmark for Naturalistic EEG Decoding

arXiv: https://arxiv.org/abs/2601.21215 · HTML: https://mlsanity.com/paper/2601.21215/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 11 references that are in the map.
- First posted on arXiv 29 Jan 2026.

## Builds on (most cited first, 6 of 6)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [Efficiently Modeling Long Sequences with Structured State Spaces](https://mlsanity.com/paper/2111.00396/) (2021)
- [Deep learning-based electroencephalography analysis: a systematic review](https://mlsanity.com/paper/1901.05498/) (2019)
- [EEG-SSM: Leveraging State-Space Model for Dementia Detection](https://mlsanity.com/paper/2407.17801/) (2024)
- [DrowzEE-G-Mamba: Leveraging EEG and State Space Models for Driver Drowsiness Detection](https://mlsanity.com/paper/2408.16145/) (2024)

## Abstract (arXiv)

We study how model architecture and temporal context interact in naturalistic EEG decoding. Using the HBN movie-watching dataset, we benchmark five architectures, CNN, LSTM, a stabilized Transformer (EEGXF), S4, and S5, on a 4-class task across segment lengths from 8s to 128s. Accuracy improves with longer context: at 64s, S5 reaches 98.7%+/-0.6 and CNN 98.3%+/-0.3, while S5 uses ~20x fewer parameters than CNN. To probe real-world robustness, we evaluate zero-shot cross-frequency shifts, cross-task OOD inputs, and leave-one-subject-out generalization. S5 achieves stronger cross-subject accuracy but makes over-confident errors on OOD tasks; EEGXF is more conservative and stable under frequency shifts, though less calibrated in-distribution. These results reveal a practical efficiency-robustness trade-off: S5 for parameter-efficient peak accuracy; EEGXF when robustness and conservative uncertainty are critical.