— Albert Einstein
2025
Read — Albert Einstein (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-10973.txt.
Introduction
Shuchen Wu's thesis asks how minds carve raw sensation into reusable parts in a 2025 dissertation. Perception arrives as a high-dimensional stream, yet cognition swiftly finds chunks within it and relations among them. Wu treats a sensory stream as a sequence and studies chunking, the split of sequences into learned parts. Reaction-time experiments showed people adapt to hidden chunks while trading speed against accuracy.
Wu then built models that learn chunks and read sequences chunk by chunk. Chunking is framed as a rational way to find repeating patterns and nested layers. Learned chunks become building blocks for transfer and mental simulation. A second part moves to abstract patterns, sorting motif types and testing their role in memory. A learning model that adds abstract variables matched human behavior and stood apart from large language models. The thesis offers chunking and abstraction as twin principles of structured learning.
Catalog record
| YEAR | 2025 |
|---|---|
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 143 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-10973