Unified Multimodal Brain Decoding via Cross-Subject Soft-ROI Fusion
Xuanyu Hu
TL;DR
This work tackles cross-subject generalization and interpretability in multimodal brain decoding, introducing BrainROI, a cross-subject soft-ROI fusion encoder, voxel-gate fusion, interpretable prompt optimization via a locally deployed LLM, and parameterized constrained decoding. The method maps fMRI signals into a shared visual-semantic space and generates captions with a frozen MLLM, achieving leading performance on the NSD brain-captioning benchmark with notable gains in BLEU-4 and CIDEr over strong baselines. Ablation studies validate the contributions of the multi-atlas soft-ROI fusion, voxel-wise gating, and interpretable prompt optimization to cross-subject robustness and caption quality. The approach advances robust, interpretable, and auditable brain decoding with practical implications for neuroscience and AI alignment, and the authors release code and configurations to support reproduction.
Abstract
Multimodal brain decoding aims to reconstruct semantic information that is consistent with visual stimuli from brain activity signals such as fMRI, and then generate readable natural language descriptions. However, multimodal brain decoding still faces key challenges in cross-subject generalization and interpretability. We propose a BrainROI model and achieve leading-level results in brain-captioning evaluation on the NSD dataset. Under the cross-subject setting, compared with recent state-of-the-art methods and representative baselines, metrics such as BLEU-4 and CIDEr show clear improvements. Firstly, to address the heterogeneity of functional brain topology across subjects, we design a new fMRI encoder. We use multi-atlas soft functional parcellations (soft-ROI) as a shared space. We extend the discrete ROI Concatenation strategy in MINDLLM to a voxel-wise gated fusion mechanism (Voxel-gate). We also ensure consistent ROI mapping through global label alignment, which enhances cross-subject transferability. Secondly, to overcome the limitations of manual and black-box prompting methods in stability and transparency, we introduce an interpretable prompt optimization process. In a small-sample closed loop, we use a locally deployed Qwen model to iteratively generate and select human-readable prompts. This process improves the stability of prompt design and preserves an auditable optimization trajectory. Finally, we impose parameterized decoding constraints during inference to further improve the stability and quality of the generated descriptions.