Research map / Vision-language and multimodal / Human and embodied modeling
A Cognitive Process-Inspired Architecture for Subject-Agnostic Brain Visual Decoding
Jingyu Lu, Haonan Wang, Qixiang ZHANG, Xiaomeng Li
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 |
|---|---|---|---|
| use VCFlow instead of GLFA* in subject-agnostic setting | fMRI-to-video decoding · cc2017 average | frame 50-way accuracy: 9.6% -> 14.0% | Improved |
| use VCFlow instead of subject-agnostic NEURONS* | fMRI-to-video decoding · cc2017 average | video 50-way accuracy: 16.1% -> 18.2% | Improved |
| use VCFlow (subject-agnostic) instead of subject-specific NEURONS | fMRI-to-video decoding · cc2017 subject 1 | frame 50-way accuracy: 20.6% -> 14.2% | Hurt |
| add image-pretraining to brain model (ablation) | VCFlow ablation · cc2017 subject 1 | video 50-way accuracy: 12.6% -> 15.3% | Improved |
| add HCAM to pretrained brain model (ablation) | VCFlow ablation · cc2017 subject 1 | frame SSIM: 0.382 -> 0.357 | Hurt |
| add SARA after HCAM (ablation) | VCFlow ablation · cc2017 subject 1 | CLIP-pcc: 0.919 -> 0.934 | Improved |
| add HED after SARA (ablation) | VCFlow ablation · cc2017 subject 1 | video 50-way accuracy: 15.2% -> 18.9% | Improved |
| add subject classification loss L_subj to SARA (ablation) | VCFlow · cc2017 subject 1 | frame 50-way accuracy: 7.52% -> 10.7% | Improved |
| add inter-subject contrastive loss L_generic to SARA (ablation) | VCFlow · cc2017 subject 1 | frame 50-way accuracy: 10.7% -> 14.2% | Improved |
| add motion loss to HED (ablation) | VCFlow · cc2017 subject 1 | frame 50-way accuracy: 13.2% -> 11.3% | Mixed |
| add classification loss to caption-only HED (ablation) | VCFlow · cc2017 subject 1 | frame 50-way accuracy: 10.0% -> 8.2% | Hurt |
Closest papers by meaning
- See Through Their Minds: Learning Transferable Brain Decoding Models from Cross-Subject fMRI (AAAI 2025 · similarity 0.90)
- $i$MIND: Insightful Multi-subject Invariant Neural Decoding (NeurIPS 2025 · similarity 0.88)
- ZEBRA: Towards Zero-Shot Cross-Subject Generalization for Universal Brain Visual Decoding (NeurIPS 2025 · similarity 0.88)
- MoRE-Brain: Routed Mixture of Experts for Interpretable and Generalizable Cross-Subject fMRI Visual Decoding (NeurIPS 2025 · similarity 0.88)
- Toward Generalizing Visual Brain Decoding to Unseen Subjects (ICLR 2025 · similarity 0.88)
- CLIP-MUSED: CLIP-Guided Multi-Subject Visual Neural Information Semantic Decoding (ICLR 2024 · similarity 0.87)
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.
