Sign in

Patrick J. Burns

@diyclassics.bsky.social
701 followers 285 following 118 posts

Associate Research Scholar, Digital Projects @ ISAW Library | prev. Quantitative Criticism Lab (UT-Austin) & Culture, Cognition, Coevolution Lab (Harvard) | Fordham PhD, #Classics | #DigiClass, #Latin, #Greek | LatinCy dev, CLTK contrib | #Python

PostsRepliesMedia
Patrick J. Burns @diyclassics.bsky.social · 06/08/2026
This is good—and I see a lot of the recent super modular/component-based LatinCy dev throughout, with ideas of how to make it all work all together even better...
161
Patrick J. Burns @diyclassics.bsky.social · 04/08/2026
That’s great—and let me know if you notice anything that would make this more useful etc.
010
Patrick J. Burns @diyclassics.bsky.social · 04/08/2026
✨Version bump✨ latincy-preprocess v0.4.0 Added Ancient Greek preprocessing features (based on @jtauber.com excellent greek-normalisation work!) github.com/latincy/lati...
Header text for LatinCy Preprocess (v0.4.0) GitHub page, including badges. Text reads: Latin text preprocessing: U/V normalization, long-s OCR correction, diacritics stripping, macron removal, and Beta Code → Unicode Greek conversion — plus Ancient Greek elision/accent normalization — with optional Rust acceleration and spaCy integration.
093
Reposted by Patrick J. Burns
Patrick J. Burns @diyclassics.bsky.social · 02/08/2026
What should I curate/count/aggregate/visualize next in the Iliad & Odyssey?
112
Reposted by Patrick J. Burns
Patrick J. Burns @diyclassics.bsky.social · 31/07/2026
📝 New post, new series 📝 "Homeric Hapax Tables in Kumpf 1984" Quantified Homer post on a data/code reconstruction of the four summary tables in Kumpf’s 1984 book »Four Indices of the Homeric Hapax Legomena« exploratoryphilology.org/posts/kumpf-1984/ #digiclass
First page of Kumpf's Index I: An Alphabetical Listing of All the Homeric Hapax Legomena with some alpha entries
031
Patrick J. Burns @diyclassics.bsky.social · 02/08/2026
What should I curate/count/aggregate/visualize next in the Iliad & Odyssey?
112
Patrick J. Burns @diyclassics.bsky.social · 01/08/2026
tl;dr—Is ἑτέρως a Homeric hapax? Is Ἀστύνοος a hapax? “Hapactic edge cases abound.”
000
Patrick J. Burns @diyclassics.bsky.social · 31/07/2026
📝 New post, new series 📝 "Homeric Hapax Tables in Kumpf 1984" Quantified Homer post on a data/code reconstruction of the four summary tables in Kumpf’s 1984 book »Four Indices of the Homeric Hapax Legomena« exploratoryphilology.org/posts/kumpf-1984/ #digiclass
First page of Kumpf's Index I: An Alphabetical Listing of All the Homeric Hapax Legomena with some alpha entries
031
Patrick J. Burns @diyclassics.bsky.social · 31/07/2026
Back from the “Epic through Numbers” panel at CCC in Maynooth last week (and I’ve seen a movie!)—think I might start up a series of posts on… Quantified Homer…
140
Patrick J. Burns @diyclassics.bsky.social · 30/07/2026
Updated the latincy-vocab package to cover issues raised in the thread. Also put together this demo site to quickly test list building: dashboard.exploratoryphilology.org/vocab
dashboard.exploratoryphilology.org
Vocab list — LatinCy demos
LatinCy demos — Vocab list: Latin NLP with the la_core_web_lg model.
120
Patrick J. Burns @diyclassics.bsky.social · 28/07/2026
Thanks for going step-by-step through this—really helpful to better understanding different configs/setups and what works (or doesn't)... let me look into this and see what I can figure out
100
Patrick J. Burns @diyclassics.bsky.social · 28/07/2026
With the latincy pipelines, the NER (and soon NEL!) components are getting better and should increasingly fill that gap
010
Patrick J. Burns @diyclassics.bsky.social · 28/07/2026
That’s cool!—didn’t realize you had a WW version running. Fwiw my interest with this WW project is to combine the lexicon with NLP annotations as a first step toward a word sense disambiguation component for the pipelines. Making good progress there which I will write up soon
110
Patrick J. Burns @diyclassics.bsky.social · 28/07/2026
‘Musa’ or ‘musa’ (or both)—if ‘Musa’, working on some proper noun handling that might be getting in the way. Can definitely look into it either way.
110
Patrick J. Burns @diyclassics.bsky.social · 28/07/2026
That’s great—and always happy to look into anything that’s not working or doesn’t look right
011
Reposted by Patrick J. Burns
Patrick J. Burns @diyclassics.bsky.social · 27/07/2026
Why not rely on an LLM for Latin vocabulary list generation? Some thoughts here on explainable/interpretable (computational) philology... exploratoryphilology.org/posts/introd...
exploratoryphilology.org
Introducing… LatinCy Vocab – Exploratory Philology Blog
Introducing LatinCy Vocab, a Python package and spaCy component that turns plaintext Latin into pedagogically-aware, well-formatted vocabulary lists.
0133
Patrick J. Burns @diyclassics.bsky.social · 27/07/2026
Why not rely on an LLM for Latin vocabulary list generation? Some thoughts here on explainable/interpretable (computational) philology... exploratoryphilology.org/posts/introd...
exploratoryphilology.org
Introducing… LatinCy Vocab – Exploratory Philology Blog
Introducing LatinCy Vocab, a Python package and spaCy component that turns plaintext Latin into pedagogically-aware, well-formatted vocabulary lists.
0133
Reposted by Patrick J. Burns
Patrick J. Burns @diyclassics.bsky.social · 26/07/2026
📝 New post 📝 "Introducing... LatinCy Vocab" Background on a Python package and spaCy component that turns plaintext Latin into pedagogically-aware, well-formatted vocabulary lists exploratoryphilology.org/posts/introd... #digiclass #teachlatin
exploratoryphilology.org
Introducing… LatinCy Vocab – Exploratory Philology Blog
Introducing LatinCy Vocab, a Python package and spaCy component that turns plaintext Latin into pedagogically-aware, well-formatted vocabulary lists.
244
Patrick J. Burns @diyclassics.bsky.social · 26/07/2026
Wrote this post as well with motivations, etc. exploratoryphilology.org/posts/introd...
exploratoryphilology.org
Introducing… LatinCy Vocab – Exploratory Philology Blog
Introducing LatinCy Vocab, a Python package and spaCy component that turns plaintext Latin into pedagogically-aware, well-formatted vocabulary lists.
130
Patrick J. Burns @diyclassics.bsky.social · 26/07/2026
📝 New post 📝 "Introducing... LatinCy Vocab" Background on a Python package and spaCy component that turns plaintext Latin into pedagogically-aware, well-formatted vocabulary lists exploratoryphilology.org/posts/introd... #digiclass #teachlatin
exploratoryphilology.org
Introducing… LatinCy Vocab – Exploratory Philology Blog
Introducing LatinCy Vocab, a Python package and spaCy component that turns plaintext Latin into pedagogically-aware, well-formatted vocabulary lists.
244
Patrick J. Burns @diyclassics.bsky.social · 26/07/2026
Here is a screenshot with one example of possible output from the repo's demo notebook: github.com/latincy/lati....
Code to produce a formatted vocabulary list for a Latin passage using latincy-vocab
110
Reposted by Patrick J. Burns
Patrick J. Burns @diyclassics.bsky.social · 24/07/2026
🚀 New release 🚀 latincy-vocab v0.1.0 Build well-formatted Latin vocabulary lists from plaintext github.com/latincy/lati... #digiclass #teachlatin
Logo and badge information for LatinCy Vocab v0.1.0
1116
Patrick J. Burns @diyclassics.bsky.social · 24/07/2026
🚀 New release 🚀 latincy-vocab v0.1.0 Build well-formatted Latin vocabulary lists from plaintext github.com/latincy/lati... #digiclass #teachlatin
Logo and badge information for LatinCy Vocab v0.1.0
1116
Patrick J. Burns @diyclassics.bsky.social · 10/07/2026
✨Version bump✨ latincy-lexicon v0.9.0 Whitaker's Words as LatinCy component Improved handling of archaic & rare forms in both lookups and paradigms, fixes to irregular verbs, better future tense handling, beta Lewis & Short integration, etc. github.com/latincy/lati...
LatinCy Lexicon logo, badge, description for v0.9.0
040
Patrick J. Burns @diyclassics.bsky.social · 09/07/2026
✨Version bump✨ txtdown v0.3.1 Minimal markup for Latin text collections Improved document validation; improved quotation handling; hierarchy-aware citations github.com/diyclassics/...
Release details for txtdown v0.3.1
0104
Reposted by Patrick J. Burns
Sebastian Heath @sebhth.bsky.social · 06/07/2026
Late to posting this job: "Research Software Engineer in Data Science and Cultural Heritage". apply.interfolio.com/188823 And while it's not @isawnyu.bsky.social specific, we do have a great DH/Digital Archaeology community here and I look forward to welcoming whomever takes this position to NYU.
042
Patrick J. Burns @diyclassics.bsky.social · 21/05/2026
Very cool—please be in touch about where things could be improved, always looking to make the models/components better.
040
Patrick J. Burns @diyclassics.bsky.social · 21/05/2026
Teaching a course this fall at @isawnyu.bsky.social on text analysis for historical languages—course description here: diyclassics.github.io/isaw-f2026-g.... If you are at NYU (or a consortium program) and are interested, let me know. #TeachAncient #DigiClass
diyclassics.github.io
Text Analysis for Historical Language Research
Graduate course at ISAW (Fall 2026) introducing computational methods for historical-language research: Python-based text analysis & NLP via word embeddings, transformer models, and large language mod...
161
Patrick J. Burns @diyclassics.bsky.social · 20/04/2026
Made some LatinCy-based suggestions here—what else are people working with?
111
Patrick J. Burns @diyclassics.bsky.social · 20/04/2026
Cool—follow up with the results here, looking forward to seeing what you find. Hope to do a good amount of multilingual in the coming year (though tbh Latin~Greek to start)
010
Patrick J. Burns @diyclassics.bsky.social · 20/04/2026
Also, latincy-readers can now build a sentence-level vector index for collections—a newer feature and one that could use more real-world testing. But if your texts are plaintext or TEI or another common format, it shouldn’t be too hard to set up. Some examples here: github.com//latincy-rea...
github.com
000
Patrick J. Burns @diyclassics.bsky.social · 20/04/2026
Maybe more immediately useful—you can get contextual token vectors and span vectors (so that includes sents) using the LatinCy pipelines on your texts, cf. this chapter and the following latincy.github.io/latincy-book...
latincy.github.io
10  Vectors – Getting Started with LatinCy
000
Patrick J. Burns @diyclassics.bsky.social · 20/04/2026
What is the application? There are w2v and floret models in the LatinCy HF repo and I have other static models(ft, glove) that could be published if helpful. Refactored LatinBERT is there too; cf. huggingface.co/latincy/mode...
huggingface.co
latincy (LatinCy)
Latin, natural language processing
310
Reposted by Patrick J. Burns
Patrick J. Burns @diyclassics.bsky.social · 14/04/2026
Announcing—LatinCy Lexicon v0.1, a refactored version of Whitaker's Words that uses LatinCy annotations to disambiguate words/meanings. Can be added as a custom component to any LatinCy pipeline. github.com/latincy/lati... #digiclass #nlproc
LatinCy Lexicon logo
3218
Patrick J. Burns @diyclassics.bsky.social · 14/04/2026
From Whitaker's release notes... "Permission is hereby freely given for any and all use of program and data." Amazingly open license, allowing us to build cool stuff for Latin. And thanks also to Martin Keegan for hosting the maintenance of Words since 2015, cf. mk270.github.io/whitakers-wo...
mk270.github.io
Whitaker's Words: Introduction
William Whitaker's WORDS, a Latin dictionary
050
Patrick J. Burns @diyclassics.bsky.social · 14/04/2026
Also—hoping to implement genuine word sense disambiguation soon by combining this functionality with the LatinCy token vectors... getting there...
010
Patrick J. Burns @diyclassics.bsky.social · 14/04/2026
We can also "reinflect" Latin forms based on existing contextual morph annotations, e.g.
Code to generate "reinflected" forms on Latin words in context with LatinCy Lexicon
010
Patrick J. Burns @diyclassics.bsky.social · 14/04/2026
Not only can we start using LatinCy annotations to disambiguate words, we can also use the WW word formation logic to generate paradigms for spaCy tokens...
Code to generate all forms of the Latin verb `amo` in the present indicative active from a single form.
120
Patrick J. Burns @diyclassics.bsky.social · 14/04/2026
Demo notebook for LatinCy Lexicon is here... github.com/latincy/lati...
github.com
latincy-lexicon/notebooks/latincy-lexicon-demo.ipynb at main · latincy/latincy-lexicon
Whitaker's Words lexical data as LatinCy pipeline components for Latin NLP - latincy/latincy-lexicon
110
Patrick J. Burns @diyclassics.bsky.social · 14/04/2026
Announcing—LatinCy Lexicon v0.1, a refactored version of Whitaker's Words that uses LatinCy annotations to disambiguate words/meanings. Can be added as a custom component to any LatinCy pipeline. github.com/latincy/lati... #digiclass #nlproc
LatinCy Lexicon logo
3218
Reposted by Patrick J. Burns
Patrick J. Burns @diyclassics.bsky.social · 31/03/2026
New blog post on the LatinCy v3.9 XPOS tags now up here: exploratoryphilology.org/posts/latinc... #digiclass #nlproc
exploratoryphilology.org
Updating LatinCy XPOS Tags – Exploratory Philology Blog
With v3.9, the LatinCy pipelines introduce a new set of human-readable, Latin-specific XPOS tags designed to surface useful linguistic information not easily derived from other UD annotations.
021
Patrick J. Burns @diyclassics.bsky.social · 31/03/2026
New blog post on the LatinCy v3.9 XPOS tags now up here: exploratoryphilology.org/posts/latinc... #digiclass #nlproc
exploratoryphilology.org
Updating LatinCy XPOS Tags – Exploratory Philology Blog
With v3.9, the LatinCy pipelines introduce a new set of human-readable, Latin-specific XPOS tags designed to surface useful linguistic information not easily derived from other UD annotations.
021
Reposted by Patrick J. Burns
Patrick J. Burns @diyclassics.bsky.social · 26/03/2026
✨ LatinCy v3.9 sm/md/lg/trf pipelines for SpaCy available ✨ - Improved tokenization and u/v norm - New custom Latin-specific XPOS tags - Better, more consistent lemma/morph coverage huggingface.co/latincy/la_c... #digiclass #nlproc
"Album" cover for the LatinCy v3.9 pipelines with "catus" from Gesner's 1551 Historia Animalium
0137
Patrick J. Burns @diyclassics.bsky.social · 26/03/2026
Making "album" covers for the model releases now... This one has Catus — from "De cato seu fele" in Gesner's »Historiae animalium liber primus de quadrupedibus viviparis« (1551), p. 318 books.google.com/books?id=89P...
021
Patrick J. Burns @diyclassics.bsky.social · 26/03/2026
✨ LatinCy v3.9 sm/md/lg/trf pipelines for SpaCy available ✨ - Improved tokenization and u/v norm - New custom Latin-specific XPOS tags - Better, more consistent lemma/morph coverage huggingface.co/latincy/la_c... #digiclass #nlproc
"Album" cover for the LatinCy v3.9 pipelines with "catus" from Gesner's 1551 Historia Animalium
0137
Reposted by Patrick J. Burns
David Bamman @dbamman.bsky.social · 16/03/2026
This has been a desideratum since the day we first published this model -- I'm so happy Patrick figured this out!
0114
Reposted by Patrick J. Burns
Patrick J. Burns @diyclassics.bsky.social · 16/03/2026
So many of you have asked—happy to announce... Latin BERT v1 now available on HuggingFace 📦 Model: huggingface.co/latincy/lati... 📝 Preprint: arxiv.org/abs/2009.10053 Original weights, experimental repackaging—leave issues/etc. in the HF discussions #nlproc #digiclass cc: @dbamman.bsky.social
huggingface.co
latincy/latin-bert · Hugging Face
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
13817
Patrick J. Burns @diyclassics.bsky.social · 16/03/2026
So many of you have asked—happy to announce... Latin BERT v1 now available on HuggingFace 📦 Model: huggingface.co/latincy/lati... 📝 Preprint: arxiv.org/abs/2009.10053 Original weights, experimental repackaging—leave issues/etc. in the HF discussions #nlproc #digiclass cc: @dbamman.bsky.social
huggingface.co
latincy/latin-bert · Hugging Face
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
13817
Patrick J. Burns @diyclassics.bsky.social · 12/03/2026
And tomorrow Fri. 3/13—I will give a talk in the late afternoon panel called... "Is agentic philology an oxymoron? Some thoughts on error, control, and disciplinary definition" ...covering some recent work on Latin post-OCR correction and its implications for philology #nlproc #digiclass
033
Patrick J. Burns @diyclassics.bsky.social · 12/03/2026
Attending the »AI & the Study of Antiquity« conference @ Rutgers University today and tomorrow...
Program for AI & the Study of Antiquity at Rutgers March 12 & 13, 2026
010