Research map / Vision-language and multimodal / Human and embodied modeling
SEED: Towards More Accurate Semantic Evaluation for Visual Brain Decoding
Juhyeon Park, Peter Yongho Kim, Jiook Cha, Shinjae Yoo, Taesup Moon
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 |
|---|---|---|---|
| combine Object F1, Cap-Sim and EffNet into SEED rather than EffNet alone (ablation) | SEED · NSD meta-evaluation | Kendall tau-b: .559 -> .621 | Improved |
| add Object F1 to EffNet (ablation) | EffNet · NSD meta-evaluation | Kendall tau-b: .559 -> .602 | Improved |
| add Cap-Sim to EffNet (ablation) | EffNet · NSD meta-evaluation | Kendall tau-b: .559 -> .583 | Improved |
| add the third metric (EffNet) to Object F1 + Cap-Sim (ablation) | SEED · NSD meta-evaluation | Kendall tau-b: .566 -> .621 | Improved |
| replace fixed 82 categories with VLM-generated object categories (ablation) | Object F1 · NSD meta-evaluation | Kendall tau-b: .516 -> .473 | Hurt |
| add object size weighting (ablation) | Object F1 · NSD meta-evaluation | Kendall tau-b: .516 -> .517 | No clear effect |
| add object location weighting (ablation) | Object F1 · NSD meta-evaluation | Kendall tau-b: .516 -> .517 | No clear effect |
| add object count partial credit (ablation) | Object F1 · NSD meta-evaluation | Pearson: .708 -> .648 | Hurt |
| swap grounding model, captioner and text encoder (Yolo-World, BLIP-2, Qwen3-Embedding) (ablation) | SEED · NSD meta-evaluation | alignment with human | No clear effect |
| evaluate SEED on a different dataset and decoder (GOD, Mind-Vis) | SEED · GOD meta-evaluation | Kendall tau-b: .453 -> .477 vs EffNet | Improved |
| use VQAScore-based similarity instead of Object F1 or Cap-Sim (ablation) | semantic similarity metric · NSD meta-evaluation | Kendall tau-b: .434 (GIT VQAScore) vs .516 Object F1 | Hurt |
| use two-way identification metrics rather than correlation to EffNet | existing metrics · NSD meta-evaluation | Kendall tau-b: CLIP .368, Inception .330 vs EffNet .559 | Hurt |
Closest papers by meaning
- Multigranular Evaluation for Brain Visual Decoding (AAAI 2026 · similarity 0.90)
- $i$MIND: Insightful Multi-subject Invariant Neural Decoding (NeurIPS 2025 · similarity 0.88)
- Toward Generalizing Visual Brain Decoding to Unseen Subjects (ICLR 2025 · similarity 0.87)
- NSD-Imagery: A Benchmark Dataset for Extending fMRI Vision Decoding Methods to Mental Imagery (CVPR 2025 · similarity 0.87)
- CLIP-MUSED: CLIP-Guided Multi-Subject Visual Neural Information Semantic Decoding (ICLR 2024 · similarity 0.87)
- Brain decoding: toward real-time reconstruction of visual perception (ICLR 2024 · 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.
