Neural networks are more modular than single neurons suggest
cs.AI
Submitted: 2026-10-09
Updated: 2026-10-14
Code: https://github.com/NeuroML-Lab/Disentangling-representations
Sources
- Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
- Are Neural Nets Modular? Inspecting Functional Modularity Through Differentiable Weight Masks
- Visual representations in the human brain are aligned with large language models
- An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
- Toy Models of Superposition
- The Platonic Representation Hypothesis
- Modeling the Human Visual System: Comparative Insights from Response-Optimized and Task-Optimized Vision Models, Language Models, and different Readout Mechanisms
- Modeling the language cortex with form-independent and enriched representations of sentence meaning reveals remarkable semantic abstractness
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems