Research map / Vision-language and multimodal / Human and embodied modeling
Visual Sequence Learning in Hierarchical Prediction Networks and Primate Visual Cortex
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 block-to-block spatiotemporal blocks with 3D convLSTM instead of frame-to-frame (ablation) | HPNet · Moving-MNIST | SSIM: 0.692 -> 0.915 | Helped |
| Use block-to-block instead of frame-to-frame (ablation) | HPNet · Moving-MNIST | MSE: 89.5 -> 65.2 | Helped |
| Use block-to-block instead of frame-to-frame (ablation) | HPNet · KTH | SSIM: 0.701 -> 0.882 | Helped |
| Use block-to-block instead of frame-to-frame (ablation) | HPNet · KTH | MSE: 103.4 -> 80.3 | Helped |
| Use HPNet feature hierarchy (F-F) instead of PredNet prediction error hierarchy | HPNet F-F · Moving-MNIST | SSIM: 0.658 -> 0.692 | Helped |
| Use HPNet B-B instead of PredRNN++ | HPNet · Moving-MNIST | SSIM: 0.872 -> 0.915 | Helped |
| Use HPNet F-F instead of PredRNN++ | HPNet F-F · Moving-MNIST | SSIM: 0.872 -> 0.692 | Hurt |
| Use HPNet B-B instead of PredRNN++ | HPNet · KTH | MSE: 86.7 -> 80.3 | Helped |
| Add more cortical modules (ablation) | HPNet B-B · KTH | decoding accuracy of movement patterns: 0.26 -> 0.63 | Helped |
| Add more cortical modules (ablation) | HPNet B-B · KTH | prediction performance | Helped |
| Sparsify feedforward computation (ablation) | HPNet · training time | training time: 13% | Helped |
| Reduce movie aperture from 8 to 3 degrees (ablation) | V2 neurons · monkey V2 recordings | predictive suppression | Hurt |
Closest papers by meaning
- Improving Visual Recognition with Hyperbolical Visual Hierarchy Mapping (CVPR 2024 · similarity 0.83)
- Meta-Learning an In-Context Transformer Model of Human Higher Visual Cortex (NeurIPS 2025 · similarity 0.81)
- STSBench: A Large-Scale Dataset for Modeling Neuronal Activity in the Dorsal Stream of Primate Visual Cortex (NeurIPS 2025 · similarity 0.81)
- Sequential Modeling Enables Scalable Learning for Large Vision Models (CVPR 2024 · similarity 0.81)
- Fundamental limits of learning in sequence multi-index models and deep attention networks: high-dimensional asymptotics and sharp thresholds (ICML 2025 · similarity 0.81)
- Towards flexible perception with visual memory (ICML 2025 · similarity 0.81)
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.