Research map / Image generation / Fast diffusion sampling
MindPilot: Closed-loop Visual Stimulation Optimization for Brain Modulation with EEG-guided Diffusion
Dongyang Li, Kunpeng Xie, Mingyang Wu, Yiwei Kong, Jiahua Tang, Haoyang Qin, Chen Wei, Quanying Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| closed-loop optimization over iterations (random -> best step) | MindPilot with AlexNet proxy · THINGS-EEG2 semantic target, 10 subjects | semantic similarity score: 0.6012 -> 0.8451 | Improved |
| same, intensity similarity | MindPilot · THINGS-EEG2 | intensity similarity score: 0.9354 -> 0.9946 | Improved |
| EEG-guided vs random guidance in closed loop (ablation) | MindPilot · Closed-loop control | CLIP score: 0.5746 -> 0.7065 | Improved |
| target-image oracle guidance vs EEG guidance (ablation) | MindPilot · Offline (200 pairs) | CLIP score: 0.8464 vs 0.6580 | Hurt |
| MindPilot vs chance-level baseline | MindPilot · EEG-to-image generation | SSIM: 0.28 -> 0.35 | Improved |
| MindPilot vs ATM-S reference ceiling | MindPilot · EEG-to-image generation | CLIP: 0.76 -> 0.67 | Hurt |
| raise reward propagation beta from 0.1 to 0.9 (ablation) | interactive search · validation set | target score: 0.6586 -> 0.6224 | Hurt |
| vary alpha (direct reward weight) from 0.1 to 0.9 at beta=0.1 (ablation) | interactive search · validation set | target score: 0.6586 -> 0.6586 | No clear effect |
| rating-driven closed loop for valence | MindPilot emotion regulation · human N=10 | mean valence rating: 0.45 -> 0.60 | Improved |
| use different pretrained backbone as proxy (SYNCLR vs DINO2) (ablation) | EEG proxy model · THINGS-EEG2 | Pearson R (%): 33.70 -> 27.69 | Hurt |
Closest papers by meaning
- MindAlign: Bridging EEG, Vision, and Language for Zero-Shot Visual Decoding (NeurIPS 2026 · similarity 0.87)
- CognitionCapturer: Decoding Visual Stimuli from Human EEG Signal with Multimodal Information (AAAI 2025 · similarity 0.87)
- Visual Decoding and Reconstruction via EEG Embeddings with Guided Diffusion (NeurIPS 2024 · similarity 0.86)
- CSBrain: A Cross-scale Spatiotemporal Brain Foundation Model for EEG Decoding (NeurIPS 2025 · similarity 0.86)
- EEG2Video: Towards Decoding Dynamic Visual Perception from EEG Signals (NeurIPS 2024 · similarity 0.86)
- BrainOmni: A Brain Foundation Model for Unified EEG and MEG Signals (NeurIPS 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
