TopoLM: brain-like spatio-functional organization in a topographic language model

2024-10-15

arXiv.org

AI Summary

This work investigates the spatial mechanisms underlying functional organization of neurons in language systems—specifically, why neighboring neurons often exhibit similar response patterns. To address this, we introduce, for the first time, an explicit two-dimensional topological structure into Transformer architectures, jointly optimizing language modeling (next-token prediction) and a spatial smoothness loss (L2 regularization combined with grid adjacency constraints). This drives the spontaneous emergence of semantically interpretable functional neuron clusters. Experiments demonstrate that the learned functional clusters align closely with human language cortex regions exhibiting grammatical and semantic selectivity, and accurately predict the spatial distribution of fine-grained linguistic features—including number, tense, and argument roles. Our results reveal that a unified spatial objective can induce brain-like functional organization in language models, establishing a novel paradigm linking neural topology with linguistic computation.

Technology Category

• Natural Language Processing: (Large) Language Models
• Computer Vision: Language and Vision
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

Neurons in the brain are spatially organized such that neighbors on tissue often exhibit similar response profiles. In the human language system, experimental studies have observed clusters for syntactic and semantic categories, but the mechanisms underlying this functional organization remain unclear. Here, building on work from the vision literature, we develop TopoLM, a transformer language model with an explicit two-dimensional spatial representation of model units. By combining a next-token prediction objective with a spatial smoothness loss, representations in this model assemble into clusters that correspond to semantically interpretable groupings of text and closely match the functional organization in the brain's language system. TopoLM successfully predicts the emergence of the spatio-functional organization of a cortical language system as well as the organization of functional clusters selective for fine-grained linguistic features empirically observed in human cortex. Our results suggest that the functional organization of the human language system is driven by a unified spatial objective, and provide a functionally and spatially aligned model of language processing in the brain.

Problem

• Understanding brain-like spatial organization in language models
• Exploring mechanisms behind functional clusters in the human language system
• Predicting spatio-functional organization in cortical language systems

Innovation

• Transformer model with 2D spatial representation
• Combines next-token prediction with spatial smoothness
• Predicts brain-like functional organization in language