Papers for

digital humanities teams

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Model explains Latin text reuse word by word with little annotation

Learning Word-Level Explanations of Text Reuse in Latin Literature with Little Annotation

Abstract: Detecting an intertextual reference is only the first step; philologists also need to know how a later author reworked the source. We explain a reuse pair word by word: each word of the reusing passage is first aligned to its source word, then labeled with the operation that transformed it, such as a copy, a change of inflection, or a substitution. Such word-level operations are expensive to annotate, and only a few corpora annotated at this level exist for text reuse in Latin literature. We describe a procedure that needs only a limited number of annotations: synthetic pairs, built by applying known operations to Latin text, train a first model, and active learning chooses the pairs an annotator corrects. Among several architectures, a model that links and labels together performs best: a pointer network that, for every reuse word, chooses a source word and names the operation in the same step. We show that with active learning, 200 pairs chosen by uncertainty are enough for the model to match one trained on every annotated pair of the 1,490 references of the Loci Similes benchmark. We release our code, word-level predictions for all references, and an interactive demo.

Fri 13 NovComputation and Language
The gist
Detecting when one Latin author copies another is just a start; understanding exactly how they changed the words helps scholars study these texts in detail. The authors created a method that needs very few hand-labeled examples to learn how each word was transformed, like copying or changing grammar. They built a model that picks which source word corresponds to each reused word and labels the change in the same step. Their approach matches results that would normally require many more annotations and comes with tools to try it out.
Open → 2611.18319v1

Annotated medieval French medical texts created for language study

ALMeD: Medieval French Gold-Standard

Abstract: Medieval French is one of the Romance vernacular languages with a significant history of transmission and an enormous influence on other vernacular languages of the time. The sources convey valuable knowledge about all aspects of culture. The study of Medieval French has, over the last few years, increasingly benefited from NLP-based methods. There have been numerous efforts to parse medieval French; however, these efforts use corpora comprising legal texts and -- primarily -- literary texts as their foundation and training data. This means that, except for juridical resources, scientific literature is overlooked. We present ALMeD, an annotated, semantically rich corpus (work-in-progress) of medical and surgical treatises produced in medieval French, i.e., Old and Middle French, including Anglo-Norman. Our semantically disambiguated gold standard goes beyond existing resources, focusing on medical terminology, and accounts for morphological development, evolving syntactical structures and the lack of spelling normalisation in these languages.

Wed 11 NovComputation and Language
The gist
People have studied medieval French texts mainly from law and literature, but medical writings have been overlooked. The authors created a carefully annotated collection of old medical and surgical texts in medieval French. Their collection highlights medical terms and language changes over time, helping computers understand these historical texts better. This resource could improve how old French medical writings are analyzed using language technology.
Open → 2611.15661v1

Hidden Markov model improves Ottoman poetry meter finding accuracy

A Lattice-Based HMM Approach to Aruz Scansion of Ottoman Poetry

Abstract: Ottoman Divan poetry is composed in aruz, a quantitative meter based on the weights of syllables. The meter is sensitive to specific characters recorded in the original writing in the Arabic script. This work is concerned with automatic meter finding and scansion alignment for Ottoman Divan poetry, especially on modern Turkish spelling without the special diacritics that scholarly transliterations of Ottoman preserve. We define two tasks on this input, first identifying the meter used in the poem, and second aligning the meter's weights with the syllabification of the poem, identifying the flaws and freedoms the poet has taken along the way. We build a corpus of 52,000 lines split by poet, with scansion annotations for 500 test lines, and a robustness benchmark generated at the deviation rates measured on real web and OCR copies. Our model is a hidden Markov model over the positions of each meter, decoded over a lattice of every candidate path, and trained by EM from the printed meter, without scansion labels. On a held-out set it scans 77.6% of the test lines fully correctly, against 26.8% for the available rule-based scanner, 55.0% for a neural tagger trained with the same supervision and 63.6% for Claude Opus 5.5 few-shot. The code is open-sourced and the tool is released on the web.

Mon 9 NovComputation and Language
The gist
Ottoman poetry uses a special rhythm based on the length of syllables, but this is hard to detect automatically, especially in modern Turkish writing without special markers. The authors created a computer method that guesses the poem's rhythm pattern and matches it to the poem's syllables, even when the writing has errors or missing details. They trained their system on many lines of poetry and tested it against other programs, finding it much more accurate. Their tool and code are also available online for others to use.
Open → 2611.12691v1

Automated system dates ancient Babylonian sky records accurately

Automated, auditable dating of the Babylonian Astronomical Diaries with language models and astronomical tools

Abstract: The Babylonian Astronomical Diaries record the sky over nearly six centuries. Many tablets are broken and have lost their date formula. Experts dated many of them one tablet at a time in the standard edition, Astronomical Diaries and Related Texts from Babylonia (ADART). They used the recorded sky and other evidence, such as year formulas and royal names. Our system automates the astronomical part for all tablets with the same rules. A rule-based parser turns the recorded observations into clauses, short statements with fixed fields. A language model checks them against the text and adds observations that the parser missed. Astronomical tools test the clauses in all 701 years from 626 BCE to AD 75. The system ranks the years by how well they agree with the clauses. The system withholds published dates until its results are final, and the output lists the passages and computed positions behind each candidate year. On 120 dated tablets, the years tied at the top of the ranking include the published year in 96% of cases. For 86 of them, this top group is a single year, the published year in 84. In 119 of the 120, the reported candidate set contains it. For 56 undated entries of ADART IV, the system gives ranked candidate years. For each tablet, the results show how far astronomy alone narrows the year. An interactive page at https://tingyuansen.github.io/Babylonian_Dating/ follows each tablet and links the code, protocol and results.

Sun 25 OctArtificial IntelligenceComputation and Language
The gist
Babylonian Astronomical Diaries are ancient texts recording sky observations for hundreds of years, but many are broken and their dates lost. The authors created a system that reads these sky observations using language models and astronomical tools to guess the correct year for each tablet. Their automated method matches expert dates in most cases and offers ranked possible dates for those without known dates. The results and tools are available online for exploration.
Open → 2610.34297v1

Suite of tools supports managing and exploring aligned multilingual texts

ParaTaxe: a Suite of Tools for Processing and Publishing Multi-parallel Corpora

Abstract: Over the past fifteen years, digital tools for parallelizing and aligning textual corpora have developed significantly within the field of digital humanities. Several projects and software tools have been created to compare different translations or versions of texts. Despite these technical advances, many parallel corpora produced in academic contexts quickly become inaccessible online due to the lack of maintenance or funding. To address this gap, we present a suite of open-source tools developed within the ACR Grimm Tradalign and ParaTaxe projects. This suite includes the AIlign aligner, based on multilingual embeddings, a Python processing pipeline for converting and aligning corpora, a web interface called ParaTaxe for managing collections of parallel corpora, and a web interface for exploring aligned corpora built with TEI-Publisher.

Fri 23 OctComputation and Language
The gist
It is hard to keep collections of linked texts in many languages available online because of limited maintenance and funding. The authors created a set of open-source tools to help process, align, and view these multilingual text collections. Their tools include a smart aligner that uses language understanding models, a processing pipeline, and web interfaces for managing and exploring the aligned texts. This approach aims to make parallel text corpora easier to maintain and access over time.
Open → 2610.32796v1

Interactive system finds shared meanings across myth and religion

MythoSemantic: An Interactive System for Discovering Semantic Parallels Across Mythological and Religious Traditions

Abstract: We present a tool for the comparative study of mythological and religious texts. Traditional manual analysis is highly labor-intensive and difficult to scale. To address this, we use neural language models to embed texts into a shared semantic space, so that users can find semantic parallels across cultural traditions in an interactive interface. Embedded fragments form distinct semantic structures whose coarse geometry aligns with well-documented historical relations between traditions. We quantify how much of that geometry is already explained by lexical overlap and by proper names. The current system is available at https://mythoscope.io, and the installable package is available at https://github.com/haafus/MythoSemantic/blob/main/RUN.md.

Tue 20 OctComputation and Language
The gist
Analyzing myths and religious texts by hand is slow and hard to do at large scale. The authors created a tool that uses AI language models to convert different texts into a shared meaning space, helping users spot similar ideas across cultures. The tool's organization of text pieces reflects known historical connections, and the authors also measured how names and word overlap explain these relationships. They provide a website and code for others to try this system.
Open → 2610.28452v1

MITRA lexicon links Sanskrit words to English Tibetan and Chinese translations

MITRA-lexicon: Lexica of attested translations from Sanskrit into English, Tibetan and Chinese

Abstract: The standard Sanskrit lexica are monumental and indispensable, but they cite their attestations as abbreviated references to editions, without links to the digitized texts now widely available and without a translation of the passage, and they naturally do not reflect translations made after them. To address these gaps, we present MITRA-lexicon, a family of corpus-grounded lexica based on span annotations of translation equivalents with Sanskrit as the pivot: Sanskrit-English, derived from the Mitrasamgraha corpus, as well as Sanskrit-Tibetan and Sanskrit-Chinese, derived from the MITRA-parallel corpus of Buddhist literature. In total, the resource comprises 2.18 million headword entries in six directional lexica. Headwords are lemmas and compounds from silver annotations that follow the Digital Corpus of Sanskrit (DCS). Senses are ranked by frequency, and each of the 10.4 million attestations includes its sentence pair, with character offsets of the word and its equivalent. We propose a multi-step pipeline based on grammatical annotation models, fine-tuned large language models, and a teacher-student setup to produce the alignments. We evaluate the lexica against reference dictionaries and manually corrected gold alignments. We release the lexica and the fine-tuned span annotation model, together with a web interface for looking up headwords and their attested senses.

Mon 19 OctComputation and LanguageMachine Learning
The gist
Standard Sanskrit dictionaries list where words appear but don’t show their actual translations or link to digital texts. The authors created MITRA-lexicon, a large digital dictionary connecting Sanskrit words with their real translated equivalents in English, Tibetan, and Chinese. This lexicon is based on millions of examples from Buddhist texts, showing words in context alongside their translations. The authors developed a system using language models to align these translations accurately and provided an online tool to explore them.
Open → 2610.25934v1

Science fiction language shifts gender words more than general fiction

Crossing the Gender Binary: Modelling Gendered Reference in Science Fiction and General Fiction

Abstract: Science fiction is widely credited with unsettling gender. This article asks whether that holds at the linguistic level, in how readily gendered reference gives way to its counterpart, a neutral term, or other social category, and whether it separates the genre from general fiction. It does so by masking man/men, woman/women, and person/people across same-period science fiction and general fiction corpora and classifying RoBERTa's five likeliest substitutes by gender. Retention and displacement are permutation-tested over five resampling rounds, and the underlying sentences are read closely. Both corpora cross the gender binary in both directions, but unequally: where woman is masked, the model predicts a masculine term about twice as often as it predicts a feminine term where man is masked. Each mismatch marks a statistical surprise, a word the corpus did not expect there. Against this shared pattern, science fiction shows greater gender permeability than the general fiction corpus, placing man in feminine contexts more often than general fiction while the reverse rate holds steady. Its degendering tendency is stronger too, drawing both gendered terms toward the neutral. Beyond the binary, a masked woman also draws substitutes naming a social category and no gender far more often than a masked man does. In measuring the gender permeability of man, woman, and person in linguistic distribution and testing whether genre changes it, the study gives science fiction studies linguistic evidence for thematic claims and computational literary studies a pipeline applicable to other category vocabularies, such as race and class.

Tue 13 OctComputation and Language
The gist
This study looks at how often words about men and women change to neutral or other gender terms in science fiction compared to regular fiction. The researchers used an AI language model to guess missing gender words in texts and checked which gendered words were predicted. They found that science fiction is more flexible with gendered words, often using neutral or different gender terms more than general fiction does. This supports the idea that science fiction explores gender in more complex ways even in everyday language.
Open → 2610.19028v1

Lüshi 1840 trains classical Chinese model with verified historical cutoff

Lüshi-1840: Building and Probing a Knowledge Cutoff for a Classical Chinese Language Model

Abstract: Language models trained only on text from before a given year are useful for studying hindsight and historical language, but the open collections of pre-modern Chinese carry no publication dates and unreliable dynasty labels. We present Lüshi-1840, a 983M-parameter model trained from scratch on 12,262 books (1.28B characters) selected as composed before 1840. A first corpus, dated by a common heuristic, in effect ended at 1644 while containing twentieth-century material. We describe the re-dating and cleaning pipeline that replaced it and an audit with markers the pipeline did not use, which still finds later material. On the model side we measure surprisal, relative to a modern reference, on annals compiled after the cutoff. For events after 1840 against events up to 1820 the ratio is 1.38 (95% interval 1.33-1.44), against about 1.00 for modern general-purpose models. In a series of 294M models, adding the later court records lowers the ratio from 1.33 to 1.16, and the two models diverge from the 1820s, where the added records begin. The probe tracks the coverage of sources and does not establish a single cutoff year; none of our probes exposes the audited contamination. Model, book list and scripts will be released.

Mon 12 OctComputation and Language
The gist
Language models that only learn from texts before a certain year help us study how language and knowledge changed over time. The authors created Lüshi-1840, a language model trained on classical Chinese books written before 1840, carefully selecting and cleaning these texts despite unreliable dates in existing collections. They tested whether the model could recognize or predict events after 1840 and found it struggles more compared to modern models, confirming it mostly learned only older language. This work shows it is possible but challenging to build language models with a true historical knowledge cutoff.
Open → 2610.16201v1

Large language models expand analysis and research workflows in cultural studies

Large Language Models, Machine-Assisted Cultural Analytics, and the Agentic Turn

Abstract: Technologies like large language models make it easier to convert cultural material into analyzable variables, while LLM-driven agentic systems extend machine assistance to other parts of the research workflow. Rather than a full how-to guide or an exhaustive review of recent LLM use cases, this chapter looks at some crucial aspects of this new paradigm as well as the bigger picture. I argue that these technologies expand the scale, methodological range, and scope for delegation, while the basic logic by which those measurements support cultural claims remains mostly unchanged.

Thu 8 OctComputers and SocietyComputation and Language
The gist
Studying culture involves turning stories, art, and history into data we can analyze. The authors explain how big AI language models help to do this faster and across more types of data without changing how we draw conclusions from that data. They also show how AI can take on more tasks in research workflows, making studying culture easier and broader. This is not a detailed guide but a discussion about what this new AI use means overall.
Open → 2610.12792v1

Character error rates affect papyrus text tasks differently

Which papyrus HTR is good enough? Character-error-rate tolerance of four papyrological tasks on Greek texts

Abstract: Purpose: Most Greek papyri remain unpublished and undigitised; a handwritten text recognition (HTR) pipeline that transcribes them automatically would let scholars discover documents and literary works that have so far gone unread. Recognition systems for Ancient Greek papyri are in statu nascendi, and how accurate they must be for a given papyrological task has not been examined. To answer this and set a benchmark for Greek papyrus HTR, we test a range of character error rates (CER) against four papyrological tasks, using published editions as ground truth. Methods: From 63,846 current editions of Greek texts in papyri.info, we imitate a letters-only "perfect HTR" output by removing the editorial layer, then degrade it with a seeded algorithm to exact CERs of 1 - 50%, with lost lines and four error-shape variants. On these data we train small models (TF-IDF, fastText, a character CNN, ByT5-small) for document type, dating and documentary-versus-literary classification, and apply eight keyword search methods. We compare models trained on clean text with models retrained at a specific CER level, and evaluate across CERs. Results: Tolerance differs by task. With clean-trained models, documentary-versus-literary classification retains 90% of its metric up to 20% CER; document type up to 7.5%; subtypes and search up to 5%; dating only up to 3%. Retraining on text containing character errors largely eliminates the sharp degradation that otherwise sets in above 15% CER. Models generally tolerate concentrated damage in a long document better than small errors spread across a short text. Conclusion: The study provides a CER target for each of the four tasks and shows that models trained on noisy text make current, imperfect text recognition useful for them.

Tue 29 SeptComputation and Language
The gist
Many ancient Greek papyrus texts are unread because they have not been digitized. The authors studied how much error in automatically reading these texts is acceptable for different research tasks. They created simulated errors in clean texts and tested how well models could still classify or search them. Some tasks like telling if texts are literary or documentary can tolerate more errors, while others like dating the texts need very low error. Training models on noisy text helps maintain accuracy even with more errors.
Open → 2609.37755v1

Computational method finds biblical references in karen blixen stories

Retrieving Biblical Intertextual References in Karen Blixen's Seven Gothic Tales

Abstract: Identifying intertextual references is central to literary scholarship, but computationally difficult when source material is transformed through paraphrase, allusion, historical language, and translation. We investigate this problem through biblical intertextuality in Karen Blixen's Seven Gothic Tales. Drawing on the commentary to a critical edition, we construct a benchmark of 189 annotated references and evaluate retrieval against all 31,170 verses of historically plausible Danish Old and New Testament translations. We compare TF-IDF and BM25 with multilingual and Danish sentence encoders, examine the effect of linguistic normalization, and fine-tune a Danish encoder using hard negatives and five-fold cross-validation. We analyze performance across automatically derived lexical-overlap strata representing quotations, paraphrases, and allusions. Linguistically normalized BM25 provides a strong zero-shot baseline, attaining an overall R@10 of 0.365 and retrieving every quotation within its ten highest-ranked verses. The best zero-shot dense model achieves a comparable overall score of 0.360 while performing better on allusions. Fine-tuning DFM-large raises its overall R@10 from 0.265 to 0.508 and more than doubles its performance on allusions, from 0.138 to 0.339. However, evaluation against editorial annotations alone understates the model's scholarly usefulness: a literary scholar judged seven of 30 selected rank-one predictions counted as false positives to be meaningful additional references. These findings show both the potential and the epistemic limits of computational intertextual retrieval. Rather than treating scholarly annotations as exhaustive or model outputs as discoveries, we propose retrieval models as heuristic co-readers that recover documented references and generate candidates for expert-led close reading.

Mon 28 SeptComputation and Language
The gist
Finding hidden references to the Bible in literature can be hard, especially when the text doesn't quote directly but hints or paraphrases. The authors worked on spotting these references in Karen Blixen's Seven Gothic Tales by comparing story passages to old Danish Bible translations. They tested different methods and improved one by training it with examples, which helped spot many more subtle references. Their approach serves as a helpful tool to suggest possible connections for scholars to explore further, rather than replacing expert reading.
Open → 2609.35765v1

Modeling chinese character evolution using manifold learning and neural ODEs

Tracing the Evolution of Oracle Bone Characters Across Three Millennia

Abstract: Of the approximately 4,500 Oracle Bone Inscription (OBI) characters discovered from the Shang dynasty, only about 1,600 have been deciphered. Many computational approaches compare OBI with glyphs from one historical period at a time. However, during the evolution of Chinese characters, significant structural or semantic changes often occur in uncertain dynasties. A single-period reference may be insufficient when relevant forms change substantially between observed eras. Therefore, we propose the \textbf{Manifold-based Script Evolution Framework (MSEF)}, a framework that models the evolution series (OBI, Bronze, Seal, Clerical, Regular) of Chinese characters as the continual evolution of a manifold space. MSEF represents each character as an era-specific manifold point and learns continuous inter-era transition rules via Neural Ordinary Differential Equations. Both manifold space and transition dynamics can be trained end-to-end through character evolution pairs across any two eras.

Mon 28 SeptComputation and Language
The gist
Many ancient Chinese characters from oracle bones are hard to understand because their forms changed a lot over thousands of years. The authors created a new method that represents each character’s shape in different historical eras as a point in a special space, tracking how those points change continuously over time. They use a kind of neural network that can model smooth transformations to learn how characters evolved from one era to the next. This helps link ancient characters with their later versions, improving understanding of their development.
Open → 2609.35674v1

Agents improve alignment of classical texts and translations with fewer defects

When Do Agents Help? Embedding, LLM and Agentic Alignment of Classical Texts and Their Translations

Abstract: Classical texts aligned with their translations support machine translation, retrieval and computational research, but evidence comparing alignment workflows is scattered. This study compares seven systems on 452 texts in Pali, Sanskrit, Mishnaic Hebrew and Tibetan, comprising 9,833 human-aligned units: four embedding pipelines, a direct LLM call, an autonomous agent, and the agent revised by an independent auditor. Generative workflows recover 93-94% of reference correspondences, against at most 77% for embeddings. A ceiling analysis shows that sentence boundaries make some references unrepresentable by the embedding pipelines. Reference recovery is similar across generative workflows: the agent's advantage is 0.5 percentage points (95% CI -0.02 to 1.17), and auditing adds no established benefit. Agents nevertheless produce structurally valid output for all 452 texts, against 437 for direct calls. A blinded three-LLM panel assesses every generative mismatch against the source and human reference. Most mismatches are labelled defensible editorial variation; consensus major-error labels cover only 0.06-0.14% of units. The panel labels significantly fewer residual defects for agents than direct calls (0.7% versus 1.4%), suggesting that reference recovery alone understates alignment quality. On ten long Pali discourses taken as published online, agents and audited agents raise recovery from the direct call's 71% to 84% and 92%. Identical reference-located chunks bring all three to 93%. Agents thus improve structural reliability and reduce judged defects on short passages, while their large recovery advantage on long documents disappears after chunking. In this setting, independent auditing offers little measurable additional benefit on prepared passages.

Sun 27 SeptComputation and Language
The gist
Matching ancient texts with their translations helps computers translate and study these works, but it is hard to do accurately. The authors compare several computer methods to line up passages in original texts with their translations. They find that AI agents can recover more matching parts than standard embedding techniques and make fewer mistakes. However, the advantage of agents is smaller when long texts are split into chunks. Having a second check by another AI adds little benefit. Overall, agents produce more reliable and better-aligned outputs.
Open → 2609.33691v1

Tracing how researchers track changing ideas in physics fields

Tracing individual knowledge trajectories in a changing field: the case of general relativity and gravitation

Abstract: Historians have reconstructed the twentieth-century transformation of general relativity and gravitation (GRG) at the field level and through individual careers, but connecting these scales requires a way to compare researchers with the changing field over time. We develop such a comparison, setting a researcher's publications and references against GRG field literature from the same, earlier, and later two-year periods. Building on Own Vocabulary and Embedding Density Estimation from our earlier two-case study (arXiv:2501.00391), we extend the analysis to the fifty most-published authors in a NASA/ADS corpus of about 180,000 GRG records (1911 to 2000) and add two citation-based measures, Referenced Vocabulary and Citation Identity. The four measures compare an author's written language, cited literature, semantic neighbourhood, and cited-authority configuration with the surrounding field. The earlier cases suggested that closer field-vocabulary alignment accompanies a denser semantic neighbourhood. Across the fifty authors this holds only partially. Written and cited vocabularies tend to move together, usually resembling later GRG literature as the field turned towards astrophysical and cosmological research. Semantic neighbourhoods more often lie where the field's publications were concentrated in earlier periods, while co-citation patterns follow no single temporal direction, and the two citation measures frequently place the same researcher differently despite drawing on identical reference lists. Individual trajectories can thus combine vocabulary tied to later field states with older semantic or citation structures, and these divergent cases mark patterns for closer historical investigation. The approach transfers to other fields with defensible corpus boundaries and adequate coverage of texts, references, and disambiguated author identities.

Wed 16 SeptComputation and Language
The gist
Understanding how scientists' work connects to the bigger field over time is hard because the field itself changes. The authors develop new ways to compare what a researcher writes and cites with the wider field of physics literature at different times. They studied 50 top authors in general relativity and gravitation and found that authors’ word choices and citations don’t always move in sync with the field’s evolution. This method helps spot unique patterns in researchers’ work and could be applied to other fields with enough text and citation data.
Open → 2609.18697v1

Luke is confirmed as author of both gospel and Acts books

Is Luke the Author of a Gospel and the Acts of the Apostles?

Abstract: According to Christian tradition, Luke is credited with authoring a Gospel and the Acts of the Apostles, even if his name does not appear in either book, both originally written in Koine Greek. Several biblical scholars assume that both texts were written by a common author, while others deduce the presence of two authors. Different studies have been found to support either finding, some based on qualitative evaluation, while a few others consider the occurrence frequency differences between the two books. To propose an enhanced quantitative analysis, this study is grounded on two recent authorship attribution models. The Burrows' Delta, applied with eleven different feature sizes, demonstrates common authorship. An author verification model confirms this finding. The following experiments consider several stylistic representations, feature sizes, and distance functions to confirm that Luke is the true author of both books.

Tue 15 SeptComputation and LanguageArtificial IntelligenceDigital Libraries
The gist
There has been debate about whether Luke wrote both a Gospel and the Acts of the Apostles because his name is not directly on these texts. Some scholars think there was one author, others think two different ones wrote them. The authors used computer-based methods to analyze writing style and vocabulary frequency in both books. Their analysis strongly supports that the same person, likely Luke, wrote both.
Open → 2609.17762v1