Tracing the Genealogies of Ideas with Large Language Model Embeddings

TL;DR

This paper introduces a new computational method using large language model embeddings to trace intellectual influence and idea evolution across extensive textual corpora, capturing semantic and structural similarities.

Contribution

The paper presents a novel ensemble approach combining semantic and structural embeddings to detect ideas and influence in large, diverse textual datasets, including 19th-century publications.

Findings

Effective detection of ideas across 400,000 texts

Capable of identifying Darwinian influence in texts

Robust to paraphrasing and structural variations

Abstract

In this paper, I present a novel method to detect intellectual influence across a large corpus. Taking advantage of the unique affordances of large language models in encoding semantic and structural meaning while remaining robust to paraphrasing, we can search for substantively similar ideas and hints of intellectual influence in a computationally efficient manner. Such a method allows us to operationalize different levels of confidence: we can allow for direct quotation, paraphrase, or speculative similarity while remaining open about the limitations of each threshold. I apply an ensemble method combining General Text Embeddings, a state-of-the-art sentence embedding method optimized to capture semantic content and an Abstract Meaning Representation graph representation designed to capture structural similarities in argumentation style and the use of metaphor. I apply this method to vectorize sentences from a corpus of roughly 400,000 nonfiction books and academic publications from the 19th century for instances of ideas and arguments appearing in Darwin's publications. This functions as an initial evaluation and proof of concept; the method is not limited to detecting Darwinian ideas but is capable of detecting similarities on a large scale in a wide range of corpora and contexts.