From Words to Worlds: Compositionality for Cognitive Architectures
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916744755412992 |
|---|---|
| author | Dhar, Ruchira Søgaard, Anders |
| author_facet | Dhar, Ruchira Søgaard, Anders |
| contents | Large language models (LLMs) are very performant connectionist systems, but do they exhibit more compositionality? More importantly, is that part of why they perform so well? We present empirical analyses across four LLM families (12 models) and three task categories, including a novel task introduced below. Our findings reveal a nuanced relationship in learning of compositional strategies by LLMs -- while scaling enhances compositional abilities, instruction tuning often has a reverse effect. Such disparity brings forth some open issues regarding the development and improvement of large language models in alignment with human cognitive capacities. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_13419 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | From Words to Worlds: Compositionality for Cognitive Architectures Dhar, Ruchira Søgaard, Anders Computation and Language Artificial Intelligence Computers and Society Machine Learning Symbolic Computation Large language models (LLMs) are very performant connectionist systems, but do they exhibit more compositionality? More importantly, is that part of why they perform so well? We present empirical analyses across four LLM families (12 models) and three task categories, including a novel task introduced below. Our findings reveal a nuanced relationship in learning of compositional strategies by LLMs -- while scaling enhances compositional abilities, instruction tuning often has a reverse effect. Such disparity brings forth some open issues regarding the development and improvement of large language models in alignment with human cognitive capacities. |
| title | From Words to Worlds: Compositionality for Cognitive Architectures |
| topic | Computation and Language Artificial Intelligence Computers and Society Machine Learning Symbolic Computation |
| url | https://arxiv.org/abs/2407.13419 |