Hilmar Lapp

Increasing participation at events promoting #OpenScience and #OpenSource in #bioinformatics: Call for the first 2026 round of the OBF Event Fellowship & overview of the overall round of 2025: open-bio.org/2026/03/05/eve…

Don't miss the Imageomics Seminar “Rethinking the Semantics of Biological Trait Recognition” with Sabina Leonelli on interoperable data, bio-ontologies, and ethics in AI science on February 23rd at 4-5PM EST! Register: go.osu.edu/sabina-leonelli

Our catalog is feature-rich but also a single-page app easy to deploy and host sustainably. It's the benefit from making datasets, models, and codebases FAIR through utilizing public repository infrastructure (GH, HF) and, importantly, their respective metadata support mechanisms

We’re recruiting! Dryad is looking for a full-time #datacurator to join our team. Fully remote, but fully supported. If you have a passion for research data quality and open data publication, then take a look: buff.ly/3qtjoIG.

Here's the thing. We have **NO IDEA** how to pick good graduate students. I served on admission committees for 10+ years, and chaired a few, and what I learned is that all the spreadsheets of grades and test scores and recommendations and essays and publications and interview rubrics are just an elaborate ruse to pretend we know what we're doing when we simply don't. Many of the most highly ranked applicants to our "top" program flamed out quickly, and tons of the students we summarily rejected have turned into amazing scientists. But in the name of creating meritocratic seeming rankings that are more about creating a workforce than great scientists (a system that anyone paying attention knows is bullshit), we've created a homogenous process adopted by nearly all institutions that has stamped out the one thing we should be striving for - given our lack of any clear understanding of what leads to success - a wide range of difference talents and experiences.

We’re excited to announce the #BeetlePalooza2024 workshop coming up in August. Apply to attend before June 10! Help shape the future of biodiversity data collection with #ai and #computervision

📸🌿Discover the future of ecological research. @BritishEcolSoc published #Imageomics workflow by Meghan Balk @hlapp and team. Their automated, reproducible science is setting new standards! go.osu.edu/balk-workflow @OSUbigdata @OSUengineering @DukeU @iDigBio @Tulane

A very useful paper by @MegBalk with @hlapp and colleagues we start to work up the image infrastructure needs for dissco-uk.org - see besjournals.onlinelibrary.wiley.com/doi/full/10.11…

Me too!!

Wow, @ProjectJupyter is the winner of the White House OSTP "Technical Advancement to Enable Open Science" category. Amazing to see recognition of the project at a national level! Makes me proud of the ecosystem of interconnected tools we've built. whitehouse.gov/ostp/news-upda…

ICYMI, some super exciting biology & medical ML models were recently released 🤩 They span from longitudinal EHR, to antibodies, to biological (taxonomy) classification, to large-scale unsupervised medical imaging models Check them out! 🧵⬇️

Introducing BioCLIP: A Vision Foundation Model for the Tree of Life imageomics.github.io/bioclip/ A foundation model that strongly generalizes on the tree of life (2M+ species), outperforming OpenAI CLIP by 18% in zero-shot classification, and supports open-ended classification over almost the entire tree of life What's the secrete ingredients? > Data: we curate and release TreeOfLife-10M, the largest and most diverse ML-ready dataset of organism images to date. It contains 10.4M images for over 450K taxa, sourced from iNaturalist, BIOSCAN, and Encyclopedia of Life. > Modeling: we creatively repurposes CLIP's multimodal contrastive learning objective for hierarchical image classification. The autoregressive language model naturally encodes the hierarchy of the tree of life taxonomy, which in turn bakes the hierarchical representation into the vision transformer encoder. Key results > Strong zero/few-shot classification for animals/plants/fungi, including rare species, outperforming CLIP by avg 16-18% absolute. > T-sne visualization shows that BioCLIP's vision encoder has captued the fine-grained hierarchical structure of the tree of life > BioCLIP is a kind of universal classifier for the tree of life. Just give it an organism image and it will likely find the correct species (among top 5)! But use it with caution; it's not perfect yet.. Final remarks > AI for Science is really hard but extremely rewarding! It took us a ton of time (1+ year) and frustration trying to find a plausible way to integrate the tree of life taxonomy into foundation model training. But when the "Eureka!" moment came and the idea hit us (by the great @weilunchao) that CLIP's multimodal contrastive learning objective can be repurposed for that, everything just follows naturally. It was truly a moment of joy and excitement! > BioCLIP is our first attempt at foundation models for biology, but it certainly won't be the last! There's so much more to do at the intersection of one of the oldest scientific disciplines and the young but thriving field of AI. Biological intelligence is the foundation for artificial intelligence, and artificial intelligence will in turn become the most important tool for us to unraval the mysteries of biological intelligence. We are hiring postdocs and PhDs in the NSF @imageomics institute to explore this exciting field! Drop us an email. also happy to chat about it at #NeurIPS2023 with any of Tanya, @weilunchao, or me. - paper: arxiv.org/abs/2311.18803 - project: imageomics.github.io/bioclip/ - demo: huggingface.co/spaces/imageom… - model: huggingface.co/imageomics/bio… - data (TreeOfLife-10M): to be released on Hugging Face soon joint work with the amazing @imageomics team: @samstevens6860 Lisa Wu, Matt Thompson, Elizabeth Campolongo @luke_ch_song @Carlyn2015 @donglixp @dahdulw Chuck Stewart, Tanya Berger-Wolf @weilunchao @ysu_nlp

1️⃣ Share it.  2️⃣ Use it.  3️⃣ Cite it 10 of our most popular datasets in this month’s Open Data Digest, and over 50,000 more at datadryad.org buff.ly/47XRkkQ #OpenScience #DataReuse #DataScience

🗣Now open 🗣 @cziscience has partnered with @KavliFoundation + @wellcometrust + @ResearchSoft to open a new funding opportunity for Essential #OpenSource Software for Science. Learn more & apply czi.co/OpenSourceSoft…

If you're an #OhioState #undergrad looking for #oncampus #employment we are hiring! Learn key skills in #ProgramManagement, #communications, #marketing and #researchadministration. Apply now! myworkday.com/osu/d/inst/15$…

What do you get when you put computer, data, and domain scientists together in a room for 3 days? Seven really cool new imageomics projects! Learn more about the recent event and all these wild new projects! lnkd.in/gZM-EMwR

RENCI at UNC Chapel Hill has an opening for a Research Scientist with expertise in integrating knowledge graphs and machine learning, and an interest in neurosymbolic methods. Here's the job posting: unc.peopleadmin.com/postings/264255 Please share!

1️⃣ Share it. 2️⃣ Use it. 3️⃣ Cite it. Discover 10 of our most popular datasets in this month’s Open Data Digest, and over 50,000 more at buff.ly/3QxssKE #OpenData #OpenScience #DataReuse #DataScience