Qihong Lu

A neural network model of free recall learns multiple memory strategies

This paper shows neural networks can develop diverse memory strategies for free recall, including an item index code similar to memory palace techniques, which outperforms temporal context models.

A neural mechanism for online discovery of latent contexts

This paper introduces NeuraGEM, a neural architecture that models online discovery of latent contexts through fast transient activity and slow synaptic plasticity, reproducing human contextual learning features and brain-like dynamics.

A neural network with key-value episodic memory retrieves and organizes memories based on causal event structures

Episodic memory supports the acquisition of structured task representations

This study shows that episodic memory stabilizes task representations, enabling structured knowledge accumulation in a neural network model.

When to retrieve and encode episodic memories: a neural network model of hippocampal-cortical interaction

This study uses a neural network model to show that selective retrieval and encoding of episodic memories improves prediction performance by reducing the risk of irrelevant memory retrieval.

Evidence for a deep, distributed and dynamic semantic code in human ventral anterior temporal cortex

This study shows semantic information in human ventral anterior temporal cortex is dynamically distributed with stable and changing elements, supporting a hub-like role in semantic networks.