How the Brain Stores Memories and Its Inspiration for Long-Context LLMs

​In this talk, we will explore, at a high level, how the brain stores memories. Building on this, we will then discuss recent advancements in AI that aim to store long-term context in large language models.

​We will also cover the following topics at a high level:

• ​What is Computational Cognitive Neuroscience (CCN)?

• ​Structure of neurons and how networks form inside the brain

• ​Hebbian learning

• ​Structure of the brain, with a closer look at the working of the hippocampus

• ​Recent advancements in brain-inspired long-context memory for AI

​About the speaker:
Yashwardhan Chaudhuri
Co-founder of ResXiv, a unified AI-powered research workspace, and former Associate ML Scientist at Wadhwani AI.
/yashwardhan-chaudhuri

To add the event to your calendar:
https://bit.ly/4pKlyQx

​Preread:

1. ​Computational Cognitive Neuroscience (Chapter 1, 2, 3, 4, 7) : https://compcogneuro.org/book

2. ​Teaching AI to Remember: Insights from Brain-Inspired Replay in Continual Learning: https://arxiv.org/abs/2509.00047

3. ​Artificial Hippocampus Networks for Efficient Long-Context Modeling: https://arxiv.org/abs/2510.07318

4. ​Reading Chapter 1 from the CCN book and Artificial Hippocampus Networks for Efficient Long-Context Modeling would be desirable.