# SABER: A Semantic-Aligned Brain Network Analysis Framework via Multi-scale Hypergraphs

arXiv: https://arxiv.org/abs/2607.01901 · HTML: https://mlsanity.com/paper/2607.01901/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 22 references that are in the map.
- First posted on arXiv 2 Jul 2026.

## Builds on (most cited first, 11 of 11)

- [Hypergraph Neural Networks](https://mlsanity.com/paper/1809.09401/) (2018)
- [Brain Network Transformer](https://mlsanity.com/paper/2210.06681/) (2022)
- [Community-Aware Transformer for Autism Prediction in fMRI Connectome](https://mlsanity.com/paper/2307.10181/) (2023)
- [MedSAM3: Delving into Segment Anything with Medical Concepts](https://mlsanity.com/paper/2511.19046/) (2025)
- [Evaluating LLM -- Generated Multimodal Diagnosis from Medical Images and Symptom Analysis](https://mlsanity.com/paper/2402.01730/) (2024)
- [BrainNPT: Pre-training of Transformer networks for brain network classification](https://mlsanity.com/paper/2305.01666/) (2023)
- [Open-vocabulary Auditory Neural Decoding Using fMRI-prompted LLM](https://mlsanity.com/paper/2405.07840/) (2024)
- [fMRI-LM: Towards a Universal Foundation Model for Language-Aligned fMRI Understanding](https://mlsanity.com/paper/2511.21760/) (2025)
- [BrainPrompt: Multi-Level Brain Prompt Enhancement for Neurological Condition Identification](https://mlsanity.com/paper/2504.16096/) (2025)
- [MADE-for-ASD: A Multi-Atlas Deep Ensemble Network for Diagnosing Autism Spectrum Disorder](https://mlsanity.com/paper/2407.07076/) (2024)
- [Aligned with LLM: a new multi-modal training paradigm for encoding fMRI activity in visual cortex](https://mlsanity.com/paper/2401.03851/) (2024)

## Abstract (arXiv)

Effective brain disease diagnosis requires the synergy of brain connectivity patterns and high-level semantic knowledge. Existing methods, however, largely treat semantics from large language models (LLMs) as auxiliary features or supervision, limiting their direct role in decision-making and constraining classification stability and robustness. To overcome this, we propose a semantic-aligned brain network framework that actively integrates LLM-derived semantics into the prediction process. Specifically, ROI-level semantics are first incorporated via global self-attention to enrich node representations and provide whole-brain context. Multi-scale hypergraphs are then constructed to explicitly model functional subnetworks and multi-ROI interactions, addressing the locality limitations of traditional GNNs and capturing high-order dependencies. Finally, a decision-level semantic alignment mechanism selectively injects patient-specific textual embeddings into graph representations, enabling semantics to directly guide predictions without perturbing the underlying network structure. Experiments on public brain network datasets ABIDE and ADHD-200 demonstrate state-of-the-art performance, enhanced stability, and improved interpretability, particularly in small-sample settings.