A small cognitive core explains human behavior at scale

arXiv:2610.13603 · cs.LG · Submitted 2026-10-09 · Read on arXiv

cs.LG

Submitted: 2026-10-09

Updated: 2026-10-09

Code: https://github.com/fla-org/flash-linearattention

Project page: https://core-cognition.github.io/core-model/results

Terminology

• Core — 1× in this paper · explained in 2 episodes

Sources

• Simple and Effective Zero-shot Cross-lingual Phoneme Recognition
• Post-training makes large language models less human-like
• Fast weight programming and linear transformers: from machine learning to neurobiology
• Gated DeltaNet-2: Decoupling Erase and Write in Linear Attention
• Universal Transformers
• DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
• HuggingFace's Transformers: State-of-the-art Natural Language Processing

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks