Tessera
Folgen

@steinly0.bsky.social

Astro Physicist. Scientific Director @arXiv Planets, Stars, Black Holes & Stuff. Whole Universes, If Needed. Dynamics of Cats; To Say Nothing of The Dog.

210 Posts155 Follower146 Folgt
@steinly0.bsky.social · 4w ago

Excessive submission rate Articles submitted to arXiv must be of original, novel, and significant self-contained research. There is a practical limit to the rate at which appropriate, independent submissions can be produced by any one person.

Richard Sever@richardsever.bsky.social · last mo.

“move from preprint to peer-reviewed publication leaves the central claims of most abstracts intact, indicating preprints are a reliable source…papers that were never posted as preprints were retracted at roughly twice the rate of those that were…” www.biorxiv.org/content/10.6...

Tracking claim changes from preprint to publication across 72,644 biomedical studies using large language models

Preprints now disseminate a large share of biomedical research before peer review. Because they have not yet passed peer review, some scientists regard preprint claims as unverified or potentially unr...

biorxiv.org

Deyan Ginev@dginev.bsky.social · last mo.

👉Still a work in progress. We see the early viability of the approach. But until it is completely stabilized, we are not "ready". Find more details in our preprint, to be presented at ICMS 26, together with a public release of the Rust code. arxiv.org/abs/2605.16562 5/5

Scaling Accessible Mathematics on arXiv: HTML Conversion and MathML 4

We report on the ongoing development of arXiv's HTML Papers offering, available on every new TeX/LaTeX submission since its initial release in 2023. The main highlights from 2025 and early 2026 are:...

arxiv.org

Deyan Ginev@dginev.bsky.social · last mo.

👉AI. The project is now heavily based on agentic coding, mostly via the Opus 4.6-4.8 series of models. The last 3 months have been shocking and transformative for me, still trying to adapt and make sense of it all. 4/5

LaTeXML: Perl source vs Rust port, weekly statistics for lines of code in the repository, over 20 years, 2006 to 2026. LaTeXML perl shows gradual growth over the period reaching 74,432 lines. latexml-oxide, the Rust port, grows in 2 phases from 2015-2019, then 2022-2024, reaching 48,000 lines of code. It then exponentiates up to 240,506 lines in the months since March 2026.
Deyan Ginev@dginev.bsky.social · last mo.

👉NYC. ar5iv's build system is newly a "Manhattan project". All machines are now collocated in NYC, with immense gratitude to the KWARC research group for two decades of support from Germany. 3/5

Deyan Ginev@dginev.bsky.social · last mo.

👉Rust. arXiv's largest month of >30,000 LaTeX documents was ar5iv's fastest conversion to HTML to date. LaTeXML now has a translation to the Rust programming language. It is nearly feature-complete and measuring at 10-30 times faster than the original. 2/5

The Open Journal of Astrophysics@ojastro.bsky.social · last mo.

New Publication at the Open Journal of Astrophysics: "A systematic survey for hypervelocity runaways from thermonuclear supernovae" by Kareem El-Badry (Caltech, USA), and 18 others based in the USA, Germany, Austria and the UK. doi.org/10.33232/001...

A systematic survey for hypervelocity runaways from thermonuclear supernovae | Published in The Open Journal of Astrophysics

By Kareem El-Badry, Klaus Werner & 17 more. A systematic survey of hypervelocity runaways, resulting from white dwarf explosions in binary systems. The findings suggest a diversity of remnant masses, ...

doi.org

@steinly0.bsky.social · last mo.

@arxiv will have a scheduled shutdown at 2 pm EDT for transition: "arXiv submission will be down for maintenance beginning 14:00 EDT Tuesday June 30th. The site should otherwise remain in operation." it might look a little different when it comes back up but no substantive changes

Egon Willigh☮gen 🟥@egonw.mastodon.social.ap.brid.gy · last mo.

maybe as fascinating about the two Max Planck papers is that they have been retracted since at least 2018 and 2015 and everyone was fine with that until that https://doi.org/10.48550/arXiv.2605.17534 preprint

The Curious Case of Max Planck retracted papers. When past scientific practices meet contemporary publishing norms

This article examines the case of two papers published in Naturwissenschaften by the physicist Max Planck that were retrospectively marked as retracted on Springer digital platform. Rather than originating in scientific fraud, these withdrawals appear to result from contemporary digitization and copyright-management procedures applied anachronistically to historical publications. Through an investigation of the circulation history of Planck 1940 and 1942 philosophical essays, the article shows that republication across multiple formats was a common and legitimate practice within the scientific publishing culture of the early 20th century. Such practices only became problematic with the later transformation of the scientific article into a countable and proprietary unit within systems of bibliometric evaluation and commercial academic publishing. This article argues that contemporary notions such as duplicate publication and self-plagiarism are historically situated categories that cannot be applied retrospectively without distorting the historical record. More broadly, the Planck case reveals how digital scholarly infrastructures controlled by large commercial publishers can limit the accessibility of the scientific past. Ironically, the original papers remain accessible today through the nonprofit digital platform Internet Archive rather than through the publisher that originally issued the journal.

arxiv.org