Research map / Vision-language and multimodal / Cross-modal learning
Alignment between Brains and AI: Evidence for Convergent Evolution across Modalities, Scales and Training Trajectories
Guobin Shen, Dongcheng Zhao, Yiting Dong, Qian Zhang, Yi Zeng
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 logarithmic fit instead of linear fit for performance vs alignment (ablation) | language models · LLM Leaderboard 2 | R^2: 0.78 -> 0.80 | Improved |
| fine-tune Llama-3.1-8B into Tulu-3 | Llama-3.1-8B · LLM Leaderboard 2 | brain alignment and performance | Improved |
| apply SimPO to gemma-2-9b | gemma-2-9b · LLM Leaderboard 2 | brain alignment and performance | Hurt |
| increase language model benchmark performance | language models · LLM Leaderboard 2 | brain alignment (Pearson r): r = 0.89 | Improved |
| increase vision model ImageNet accuracy | vision models · ImageNet | brain alignment (Pearson r): r = 0.53 | Improved |
| use transformer vision models instead of CNNs (ablation) | vision models · ImageNet | brain alignment at equal accuracy | Improved |
| train MixNet-L longer | MixNet-L · ImageNet | brain alignment vs accuracy correlation: r=0.98 | Improved |
| train Pythia models longer | Pythia · PIQA and LAMBADA | brain alignment vs task performance: r=0.73 to r=0.95 | Improved |
| increase benchmark difficulty from ImageNet to ImageNet-A (ablation) | vision models · ImageNet-A | Pearson r with brain alignment: 0.53 -> 0.46 | Hurt |
Closest papers by meaning
- Bridging Critical Gaps in Convergent Learning: How Representational Alignment Evolves Across Layers, Training, and Distribution Shifts (NeurIPS 2025 · similarity 0.92)
- Scaling and context steer LLMs along the same computational path as the human brain (NeurIPS 2025 · similarity 0.89)
- Brain-tuning Improves Generalizability and Efficiency of Brain Alignment in Speech Models (NeurIPS 2025 · similarity 0.88)
- Attention Alignment Between Humans and Vision-Language Models (NeurIPS 2026 · similarity 0.88)
- Understanding the Emergence of Multimodal Representation Alignment (ICML 2025 · similarity 0.88)
- Model–Behavior Alignment under Flexible Evaluation: When the Best-Fitting Model Isn’t the Right One (NeurIPS 2025 · 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.
