Abstract:Adaptive cognition requires structured internal models of objects and their relations. Predictive neural networks are often proposed to learn such world models, but how these are instantiated and how they support prediction remain unclear. We investigate this in a minimal in-silico setting. A recurrent neural network samples tokens sequentially from 2D continuous token scenes and is trained to predict the upcoming token from the current input and a saccade-like displacement. On novel scenes, prediction accuracy improves across the sequence, indicating in-context learning. Decoding analyses reveal path integration and dynamic binding of token identity to position. Interventional analyses show that new bindings can be learned late in sequence and that out-of-distribution bindings can be learned as well. Together, these findings show how structured representations relying on flexible binding emerge to support prediction, offering a mechanistic account of sequential world modeling relevant to cognitive science.
| Comments: | 8 pages, 4 figures; accepted at CogSci 2026 |
| Subjects: | Machine Learning (cs.LG); Neurons and Cognition (q-bio.NC) |
| Cite as: | arXiv:2602.03490 [cs.LG] |
| (or arXiv:2602.03490v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2602.03490 arXiv-issued DOI via DataCite |
From: Sushrut Thorat [view email]
[v1]
Tue, 3 Feb 2026 13:08:27 UTC (827 KB)
[v2]
Fri, 8 May 2026 10:45:31 UTC (855 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。