Saved in:
Bibliographic Details
Main Authors: Thuy, Phan Thi Thanh, Yamamoto, Akihiro
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2410.11396
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909350003474432
author Thuy, Phan Thi Thanh
Yamamoto, Akihiro
author_facet Thuy, Phan Thi Thanh
Yamamoto, Akihiro
contents In this paper we propose that a restricted version of logical inference can be implemented with self-attention networks. We are aiming at showing that LLMs (Large Language Models) constructed with transformer networks can make logical inferences. We would reveal the potential of LLMs by analyzing self-attention networks, which are main components of transformer networks. Our approach is not based on semantics of natural languages but operations of logical inference. %point of view. We show that hierarchical constructions of self-attention networks with feed forward networks (FFNs) can implement top-down derivations for a class of logical formulae. We also show bottom-up derivations are also implemented for the same class. We believe that our results show that LLMs implicitly have the power of logical inference.
format Preprint
id arxiv_https___arxiv_org_abs_2410_11396
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Implementing Derivations of Definite Logic Programs with Self-Attention Networks
Thuy, Phan Thi Thanh
Yamamoto, Akihiro
Artificial Intelligence
In this paper we propose that a restricted version of logical inference can be implemented with self-attention networks. We are aiming at showing that LLMs (Large Language Models) constructed with transformer networks can make logical inferences. We would reveal the potential of LLMs by analyzing self-attention networks, which are main components of transformer networks. Our approach is not based on semantics of natural languages but operations of logical inference. %point of view. We show that hierarchical constructions of self-attention networks with feed forward networks (FFNs) can implement top-down derivations for a class of logical formulae. We also show bottom-up derivations are also implemented for the same class. We believe that our results show that LLMs implicitly have the power of logical inference.
title Implementing Derivations of Definite Logic Programs with Self-Attention Networks
topic Artificial Intelligence
url https://arxiv.org/abs/2410.11396