Research map / Video / Video-language understanding
One Hundred Neural Networks and Brains Watching Videos: Lessons from Alignment
Christina Sartzetaki, Gemma Roig, Cees G. M. Snoek, Iris Groen
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Benchmarks about one hundred publicly available deep video models for representational alignment with human brain fMRI responses to videos, disentangling factors such as architecture, task, and training data.
Problem
Unclear what video models share with human visual processing; no large-scale benchmark of deep video models against brain fMRI responses to video existed.
Approach
First large-scale benchmarking of publicly available deep video models on representational alignment to human brain activity evoked by videos, using a recently released video fMRI dataset and disentangling contributing factors.
Experimental setup
Datasets: recently released video fMRI dataset
Benchmarks: brain alignment to video fMRI dataset
Metrics: representational alignment
Closest papers by meaning
- Bridging Critical Gaps in Convergent Learning: How Representational Alignment Evolves Across Layers, Training, and Distribution Shifts (NeurIPS 2025 · similarity 0.89)
- SIM: Surface-based fMRI Analysis for Inter-Subject Multimodal Decoding from Movie-Watching Experiments (ICLR 2025 · similarity 0.89)
- Human-Aligned Image Models Improve Visual Decoding from the Brain (ICML 2025 · similarity 0.89)
- Evaluating alignment between humans and neural network representations in image-based learning tasks (NeurIPS 2024 · similarity 0.88)
- Sparse components distinguish visual pathways & their alignment to neural networks (ICLR 2025 · similarity 0.88)
- Scaling Laws for Task-Optimized Models of the Primate Visual Ventral Stream (ICML 2025 · similarity 0.88)
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.
