mstdn.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
A general-purpose Mastodon server with a 500 character limit. All languages are welcome.

Administered by:

Server stats:

11K
active users

#reproducibility

7 posts7 participants1 post today
Christof Schöch<p>We're now in the middle of the special roundtable session on "Living With Machines: Comparative Literature, AI, and the Ethics of Digital Imagination". </p><p>I'll speak briefly about different kinds of AI (generative LLMs, non-generative LLMs, deep learning, machine learning); when to use which, and most important, when NOT to use LLMs; as well as some best practices for <a href="https://fedihum.org/tags/transparency" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>transparency</span></a>, <a href="https://fedihum.org/tags/reproducibility" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>reproducibility</span></a> and <a href="https://fedihum.org/tags/sustainability" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>sustainability</span></a> in this context. </p><p>Details: <a href="https://www.conftool.pro/icla2025/index.php?page=browseSessions&amp;cols=5&amp;form_session=1031&amp;mode=table" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">conftool.pro/icla2025/index.ph</span><span class="invisible">p?page=browseSessions&amp;cols=5&amp;form_session=1031&amp;mode=table</span></a> </p><p><a href="https://fedihum.org/tags/ICLA2025" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ICLA2025</span></a> <a href="https://fedihum.org/tags/Seoul" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Seoul</span></a> <a href="https://fedihum.org/tags/LLMs" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLMs</span></a></p>
Martin Michaelis (he/him)<p>Our latest piece on the reproducibility crisis. If we want to be serious, we will need to perform actual research to establish whether there is one and if there is, how big the problem actually is. Only with this knowledge, we can design effective measures and monitor their success. <a href="https://mastodonapp.uk/tags/reproducibility" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>reproducibility</span></a> <a href="https://mastodonapp.uk/tags/ReproducibilityCrisis" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ReproducibilityCrisis</span></a> <a href="https://mastodonapp.uk/tags/academia" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>academia</span></a> <a href="https://mastodonapp.uk/tags/research" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>research</span></a> <a href="https://mastodonapp.uk/tags/LifeSciences" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LifeSciences</span></a> </p><p><a href="https://www.tandfonline.com/doi/full/10.1080/02698595.2025.2538937" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">tandfonline.com/doi/full/10.10</span><span class="invisible">80/02698595.2025.2538937</span></a></p>
Dutch Reproducibility Network<p>Embedding <a href="https://mastodon.social/tags/reproducibility" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>reproducibility</span></a>: our annual symposium will focus on how reproducibility guidelines can be implemented in daily <a href="https://mastodon.social/tags/scientific" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>scientific</span></a> practice. </p><p>Save the date: 19 January 2026, Poortgebouw Leiden</p><p>We invited two keynote speakers and several workshop hosts. Stay tuned for program updates and the registration link!</p><p>We will also share a call for <a href="https://mastodon.social/tags/posters" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>posters</span></a> and <a href="https://mastodon.social/tags/pitches" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>pitches</span></a>. </p><p><a href="https://mastodon.social/tags/OpenScience" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>OpenScience</span></a> <a href="https://mastodon.social/tags/OpenResearch" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>OpenResearch</span></a> <a href="https://mastodon.social/tags/ResearchIntegrity" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ResearchIntegrity</span></a> <a href="https://mastodon.social/tags/CultureChange" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>CultureChange</span></a></p>
Jorge Miguel Silva<p>📢 Just published our new work on federated random forests for privacy-preserving machine learning!<br>📄 “A Federated Random Forest Solution for Secure Distributed Machine Learning”<br>📌 IEEE: <a href="https://doi.org/10.1109/CBMS65348.2025.00159" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">doi.org/10.1109/CBMS65348.2025</span><span class="invisible">.00159</span></a></p><p>📂 Supplementary slides:<br>🔗 <a href="https://doi.org/10.5281/zenodo.16539345" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="">doi.org/10.5281/zenodo.16539345</span><span class="invisible"></span></a></p><p>We're advancing secure AI without sharing data. Feedback &amp; collaborations welcome! 🚀<br><a href="https://mastodon.social/tags/FederatedLearning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>FederatedLearning</span></a> <a href="https://mastodon.social/tags/PrivacyPreservingAI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>PrivacyPreservingAI</span></a> <a href="https://mastodon.social/tags/MachineLearning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MachineLearning</span></a> <a href="https://mastodon.social/tags/OpenScience" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>OpenScience</span></a> <a href="https://mastodon.social/tags/IEEE" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>IEEE</span></a> <a href="https://mastodon.social/tags/DataScience" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>DataScience</span></a> <a href="https://mastodon.social/tags/Zenodo" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Zenodo</span></a> <a href="https://mastodon.social/tags/ResearchSoftware" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ResearchSoftware</span></a> <a href="https://mastodon.social/tags/Reproducibility" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Reproducibility</span></a></p>
Charles Tapley Hoyt<p>Most cheminformatics code that queries ChEMBL struggles with reproducibility.</p><p>chembl-downloader can help:</p><p>&gt;&gt;&gt; import chembl_downloader as cd<br>&gt;&gt;&gt; df = cd.query("""<br> SELECT chembl_id, pref_name<br> FROM molecule_dictionary<br> WHERE pref_name IS NOT NULL<br>""")</p><p>It's even sneaking its way into <span class="h-card" translate="no"><a href="https://masto.ai/@wpwalters" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>wpwalters</span></a></span> and <span class="h-card" translate="no"><a href="https://mastodon.social/@dr_greg_landrum" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>dr_greg_landrum</span></a></span> blogs :)</p><p>Code/Docs: <a href="https://github.com/cthoyt/chembl-downloader" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">github.com/cthoyt/chembl-downl</span><span class="invisible">oader</span></a></p><p>Preprint: <a href="https://arxiv.org/pdf/2507.17783" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="">arxiv.org/pdf/2507.17783</span><span class="invisible"></span></a></p><p><a href="https://scholar.social/tags/cheminformatics" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>cheminformatics</span></a> <a href="https://scholar.social/tags/chemoinformatics" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>chemoinformatics</span></a> <a href="https://scholar.social/tags/chembl" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>chembl</span></a> <a href="https://scholar.social/tags/reproducibility" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>reproducibility</span></a> <a href="https://scholar.social/tags/chemistry" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>chemistry</span></a> <a href="https://scholar.social/tags/openscience" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>openscience</span></a></p>
Pete Bachant<p>Put succinctly, Calkit's goal is to make "repro pack" READMEs unnecessary</p><p><a href="https://fediscience.org/tags/openscience" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>openscience</span></a> <a href="https://fediscience.org/tags/reproducibility" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>reproducibility</span></a></p>
Harald Klinke<p>Offene KI braucht offene Infrastruktur – auch bei der Entwicklung.<br>Das Projekt f13 auf OpenCode stellt Tools für transparente, reproduzierbare Workflows zur Verfügung – von Modelltraining bis Deployment.<br>Ein Schritt Richtung digitaler Souveränität durch offene Werkzeuge.<br><a href="https://det.social/tags/OpenSourceAI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>OpenSourceAI</span></a> <a href="https://det.social/tags/Reproducibility" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Reproducibility</span></a> <a href="https://det.social/tags/DigitaleSouver%C3%A4nit%C3%A4t" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>DigitaleSouveränität</span></a></p><p><a href="https://gitlab.opencode.de/f13" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="">gitlab.opencode.de/f13</span><span class="invisible"></span></a></p>
Doug Ortiz<p>A surprising Reddit finding: Many MLOps practitioners lack understanding of core reproducibility tools like `setup.py` and `tox.ini`. 🤔 </p><p>This impacts model reliability in production! What are your thoughts on bridging this crucial skills gap in the MLOps community? </p><p><a href="https://mastodon.social/tags/TheLoomingMlops" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>TheLoomingMlops</span></a> <a href="https://mastodon.social/tags/MLOps" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MLOps</span></a> <a href="https://mastodon.social/tags/Reproducibility" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Reproducibility</span></a> <a href="https://mastodon.social/tags/SkillsGap" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>SkillsGap</span></a> <a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a></p><p><a href="https://dougortiz.blogspot.com/2025/07/the-mlops-reproducibility-crisis-why.html" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">dougortiz.blogspot.com/2025/07</span><span class="invisible">/the-mlops-reproducibility-crisis-why.html</span></a></p>
Technology Tales<p>Docker Desktop for Statisticians revolutionises R use by creating isolated, reproducible environments. This eliminates version conflicts and simplifies setups. With Docker, you run pre-configured R containers, enabling efficient and clean analysis environments. Explore container management to enhance statistical work and ensure easy collaborative sharing. <a href="https://mstdn.social/tags/Docker" class="mention hashtag" rel="tag">#<span>Docker</span></a> <a href="https://mstdn.social/tags/Statistics" class="mention hashtag" rel="tag">#<span>Statistics</span></a> <a href="https://mstdn.social/tags/Reproducibility" class="mention hashtag" rel="tag">#<span>Reproducibility</span></a> <a href="https://mstdn.social/tags/RStats" class="mention hashtag" rel="tag">#<span>RStats</span></a> <a href="https://www.statology.org/docker-desktop-for-statisticians-running-r-in-containers/" target="_blank" rel="nofollow noopener" translate="no"><span class="invisible">https://www.</span><span class="ellipsis">statology.org/docker-desktop-f</span><span class="invisible">or-statisticians-running-r-in-containers/</span></a></p>
Pete Bachant<p>When you describe the computational methods in your paper without sharing the code and data:</p><p><a href="https://fediscience.org/tags/OpenScience" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>OpenScience</span></a> <a href="https://fediscience.org/tags/reproducibility" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>reproducibility</span></a></p>
Pete Bachant<p>One major cause of the reproducibility crisis is that the way we describe computational methods in research articles has not kept pace with the increase in complexity of the methods themselves. Human language and mathematical formulas are not adequate in many cases.</p><p><a href="https://fediscience.org/tags/OpenScience" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>OpenScience</span></a> <a href="https://fediscience.org/tags/reproducibility" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>reproducibility</span></a> <a href="https://fediscience.org/tags/opensource" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>opensource</span></a></p>
Veit Schiele 🔜 @FrOSCon<p>Karl Popper in The Logic of Scientific Discovery, 1959: ‘Non-reproducible single occurrences are of no significance to science.’</p><p>XKCD in Replication Crisis, 2025: ‘Replication Crisis Solved’<br><a href="https://xkcd.com/3117/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="">xkcd.com/3117/</span><span class="invisible"></span></a></p><p><a href="https://mastodon.social/tags/Science" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Science</span></a> <a href="https://mastodon.social/tags/Reproducibility" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Reproducibility</span></a> <a href="https://mastodon.social/tags/XKCD" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>XKCD</span></a></p>
David Philip Morgan<p>Congratulations to <span class="h-card" translate="no"><a href="https://fediscience.org/@ElenLeFoll" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>ElenLeFoll</span></a></span> and colleagues on joining the German Reproducibility Network! 🎊 </p><p>Great to see representation of the humanities in <a href="https://mastodon.world/tags/OpenScience" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>OpenScience</span></a> and <a href="https://mastodon.world/tags/Reproducibility" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Reproducibility</span></a>, lots of great progress underway there - researchers <span class="h-card" translate="no"><a href="https://xn--baw-joa.social/@unimannheim" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>unimannheim</span></a></span> should definitely check out the talks from this initiative! 💡</p><p><span class="h-card" translate="no"><a href="https://mastodon.world/@GermanRepro" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>GermanRepro</span></a></span> <span class="h-card" translate="no"><a href="https://scicomm.xyz/@ReproducibiliTeaGlobal" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>ReproducibiliTeaGlobal</span></a></span> <span class="h-card" translate="no"><a href="https://fediscience.org/@ElenLeFoll" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>ElenLeFoll</span></a></span></p>
Martin Michaelis (he/him)<p>An interesting approach towards investigating <a href="https://mastodonapp.uk/tags/reproducibility" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>reproducibility</span></a> in <a href="https://mastodonapp.uk/tags/research" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>research</span></a>. And not just doom and gloom for a change... <a href="https://mastodonapp.uk/tags/Academia" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Academia</span></a> </p><p><a href="https://www.nature.com/articles/d41586-025-02250-1" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">nature.com/articles/d41586-025</span><span class="invisible">-02250-1</span></a></p>
Pete Bachant<p>"One button" reproducibility should be the standard.</p><p><a href="https://fediscience.org/tags/OpenScience" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>OpenScience</span></a> <a href="https://fediscience.org/tags/reproducibility" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>reproducibility</span></a></p>
Daniel Hoffmann🌻<p>There is at least one subfield of biology where <a href="https://mathstodon.xyz/tags/reproducibility" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>reproducibility</span></a> is relative high: fruit fly immunology where, according to a new study, the majority of results can be reproduced. I guess that a reason may be large sample sizes. <br><a href="https://www.nature.com/articles/d41586-025-02250-1" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">nature.com/articles/d41586-025</span><span class="invisible">-02250-1</span></a></p>

Retractions and failures to replicate are signs of weak research. But they're also signs of laudable and necessary efforts to identify weak research and improve future research. The #Trump admin is systematically weaponizing these efforts to cast doubt on science as such.

"Research-integrity sleuths say their work is being ‘twisted’ to undermine science."
nature.com/articles/d41586-025

www.nature.comResearch-integrity sleuths say their work is being ‘twisted’ to undermine scienceSome sleuths fear that the business of cleaning up flawed studies is being weaponized against science itself.