High Quality Embeddings for Horn Logic Reasoning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Yifan, White, Yasir, Clark, Dean, Sanchez, Joseph, Lipsey, Jevon, Hirst, Ashely, Heflin, Jeff
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917514371399680
author Zhang, Yifan
White, Yasir
Clark, Dean
Sanchez, Joseph
Lipsey, Jevon
Hirst, Ashely
Heflin, Jeff
author_facet Zhang, Yifan
White, Yasir
Clark, Dean
Sanchez, Joseph
Lipsey, Jevon
Hirst, Ashely
Heflin, Jeff
contents Neural networks can be trained to rank the choices made by logical reasoners, resulting in more efficient searches for answers. A key step in this process is creating useful embeddings, i.e., numeric representations of logical statements. This paper introduces and evaluates several approaches to creating embeddings that result in better downstream results. We train embeddings using triplet loss, which requires examples consisting of an anchor, a positive example, and a negative example. We introduce three ideas: generating anchors that are more likely to have repeated terms, generating positive and negative examples in a way that ensures a good balance between easy, medium, and hard examples, and periodically emphasizing the hardest examples during training. We conduct several experiments to evaluate this approach, including a comparison of different embeddings across different knowledge bases, in an attempt to identify what characteristics make an embedding well-suited to a particular reasoning task.
format Preprint
id arxiv_https___arxiv_org_abs_2605_20467
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle High Quality Embeddings for Horn Logic Reasoning
Zhang, Yifan
White, Yasir
Clark, Dean
Sanchez, Joseph
Lipsey, Jevon
Hirst, Ashely
Heflin, Jeff
Artificial Intelligence
I.2.6; I.2.4
Neural networks can be trained to rank the choices made by logical reasoners, resulting in more efficient searches for answers. A key step in this process is creating useful embeddings, i.e., numeric representations of logical statements. This paper introduces and evaluates several approaches to creating embeddings that result in better downstream results. We train embeddings using triplet loss, which requires examples consisting of an anchor, a positive example, and a negative example. We introduce three ideas: generating anchors that are more likely to have repeated terms, generating positive and negative examples in a way that ensures a good balance between easy, medium, and hard examples, and periodically emphasizing the hardest examples during training. We conduct several experiments to evaluate this approach, including a comparison of different embeddings across different knowledge bases, in an attempt to identify what characteristics make an embedding well-suited to a particular reasoning task.
title High Quality Embeddings for Horn Logic Reasoning
topic Artificial Intelligence
I.2.6; I.2.4
url https://arxiv.org/abs/2605.20467