This study investigates the potential influence of Herman Melville reading on his own writings through computational semantic similarity analysis. Using documented records of books known to have been owned or read by Melville, we compare selected passages from his works with texts from his library. The methodology involves segmenting texts at both sentence level and non-overlapping 5-gram level, followed by similarity computation using BERTScore. Rather than applying fixed thresholds to determine reuse, we interpret precision, recall, and F1 scores as indicators of possible semantic alignment that may suggest literary influence. Experimental results demonstrate that the approach successfully captures expert-identified instances of similarity and highlights additional passages warranting further qualitative examination. The findings suggest that semantic similarity methods provide a useful computational framework for supporting source and influence studies in literary scholarship.
Figures & tables
Figure 1: 5-gram analysis: F1 scores for instance 1 in Table 1 comparing Typee with A visit to the south seas .
Figure 2: Sentence level analysis: F1 scores for instance 1 in Table 1 comparing Typee with A visit to the south seas .
Figure 3: Precision, recall and F1 score values for instance 4 from Table 1 comparing The Natural History of the Sperm Whale and Moby Dick
Figure 4: Heat map for complete text comparison of instance 2 in Table 1 comparing Life and Remarkable Adventures of Israel R. Potter with Israel Potter; His fifty years of exile at the sentence level.
Melville Text
Read Text
Results
Israel had now been three days without food, except one two-penny loaf. (sentence 11)
I had now been three days without food (with the exception of a single two-penny loaf) and felt myself unable much longer to resist the cravings of nature—my spirits, which until now had armed me with fortitude began to forsake me—indeed I was at this moment on the eve of despair! When, calling to mind that grief would only aggravate my calamity, I endeavoured to arm my soul with patience; and habituate myself as well as I could, to woe. Accordingly I roused my spirits; and banishing for a few moments these gloomy ideas, I began to reflect seriously on the methods by which to extricate myself from this labyrinth of horror. (sentence 6)
p = 0.80 r = 0.43 F1 = 0.56
presenting a smaller supply than (chunk 301)
which containing a small quantity (chunk 88)
p = 0.65 r = 0.70 F1 = 0.68
Table 2: Two text pair examples for model discovered parallels from Melville’s Israel Potter; His Fifty Years of Exile and Life and Remarkable Adventures of Israel R. Potter . The first is at the sentence level and the second at the 5-gram level.
Calculating semantic textual similarity is a foundational task in natural language processing. Current large language models (LLMs) based methods typically rely on extracting last-layer hidden states with fixed dimensions to compute similarity for every text pairs. We argue that this paradigm is suffer from two limitations: (i) The last hidden layer encodes more general knowledge rather than just semantic knowledge, making it suboptimal for semantic similarity computation; (ii) The hidden layer dimensions of LLMs are generally very large, which introduces some redundancy and noise for representing semantics. In this work, we propose DySem, a novel training-free framework that investigates more semantic-related internal components of LLMs via multilingual consensus, and shifts away from static representation spaces in favor of dynamic, sample-specific semantic dimensions by constructing text-dependent joint semantic set and computes similarity over this shared dimensional subset. Extensive experiments across various LLMs show that our method consistently outperforms recent baselines while maintaining lower dimensions for similarity calculation. The code is released at https://github.com/szu-tera/DySem.
Kaijie Zheng, Weiqin Wang, Yile Wang +1
College of Computer Science and Software Engineering, Shenzhen University
While digitized corpora have transformed the study of intellectual transmission, current methods rely heavily on lexical text reuse detection, capturing verbatim quotations but fundamentally missing paraphrases and complex implicit engagement. This paper evaluates semantic search in 18th-century intellectual history through the reception of John Locke's foundational work. Using expert annotation grounded in a semantic taxonomy, we examine whether an off-the-shelf semantic search pipeline can surface meaning-level correspondences overlooked by lexical methods. Our results demonstrate that semantic search retrieves substantially more implicit receptions than lexical baselines. However, linguistic diagnostics also reveal a "lexical gatekeeping" effect, where retrieval remains partially constrained by surface vocabulary overlap. These findings highlight both the potential and the limitations of semantic retrieval for analyzing the circulation of ideas in large historical corpora. The data is available at https://github.com/COMHIS/locke-sim-data.
Yu Wu, Ananth Mahadevan, Filip Ginter +2
University of Helsinki · 2TurkuNLP, University of Turku · 3ELLIS Institute Finland
We present a dual-track pipeline for detecting biblical allusions in literary fiction and apply it to the novels of Cormac McCarthy. A bottom-up embedding track uses inverse document frequency to identify rare vocabulary shared with the King James Bible, embeds occurrences in their local context for sense disambiguation, and passes candidate passage pairs through cascaded LLM review. A top-down register track asks an LLM to read McCarthy's prose undirected to any specific biblical passage for comparison, catching allusions not distinguished by word or phrase rarity. Both tracks are cross-validated by a long-context model that holds entire novels alongside the KJV in a single pass, and every finding is checked against published scholarship. Restricting attention to allusions that carry a textual echo--shared phrasing, reworked vocabulary, or transplanted cadence--and distinguishing literary allusions proper from signposted biblical references (similes naming biblical figures, characters overtly citing scripture), the pipeline surfaces 349 allusions across the corpus. Among a target set of 115 previously documented allusions retrieved through human review of the academic literature, the pipeline independently recovers 62 (54% recall), with recall varying by connection type from 30% (transformed imagery) to 80% (register collisions). We contextualise these results with respect to the value-add from LLMs as assistants to mechanical stylometric analyses, and their potential to facilitate the statistical study of intertextuality in massive literary corpora.
Ewan Cameron
Actuarial Mathematics & Statistics, School of Mathematical & Computer Sciences, Heriot-Watt University, Edinburgh EH14 4AS, UK