iiinova — where your digital self keeps evolving

Was Crick wrong? Do frozen accidents really exist? What role does chance play, and what would it take to unfreeze them? I’ve organized many conferences before, but I’m extra happy to invite you to this one 🥶. Speakers include Nick Lane, Sefy Hendler, Carolina Mangone and others (see poster).

131

From the marble Michelangelo used to sculpt David to borders between countries, past choices constrain what comes next. The choices we’re making now about AI and energy might become frozen accidents that future generations struggle to change.

130

Francis Crick controversially proposed that the genetic code was a “frozen accident”: a chance choice that became difficult to change because so much had evolved around it. On November 20 in Paris, we’ll explore this idea across biology, art and literature.

120

Oded Rechavi @odedrechavi.bsky.social · 25/09/2026

Mark your calendar! So excited to organize “Frozen Accidents” with art historian Sefy Hendler! Join us Nov 20th at Galerie Colbert (INHA), Paris. Hosted by HiCSA, Paris 1 Panthéon-Sorbonne. Please share widely, especially outside your field. Be there or be frozen!!! (see poster & thread for details)

3149

If you’re running low on fascinating biology, look no further! Leah strikes again 👇

051

Reposted by Oded Rechavi

Dan (✈️ ICML-26) @ddofer.bsky.social · 15/09/2026

Our diploid aware genomic language model is out! 1/6) The genomes we train AI models on do not exist. Most genomic language models read a single haploid sequence, but human genomes are diploid.

151

Oded Rechavi @odedrechavi.bsky.social · 15/09/2026

Departmental seminars when there's no pizza

2616

Oded Rechavi @odedrechavi.bsky.social · 02/09/2026

I'm so happy that I regained my excitement and optimism about science (owing to amazing people joining the lab)

1370

Oded Rechavi @odedrechavi.bsky.social · 06/08/2026

How it feels to leave for a short vacation

2241

Oded Rechavi @odedrechavi.bsky.social · 29/07/2026

Celine Dion's All By Myself is a strong Walk Up Song

150

Oded Rechavi @odedrechavi.bsky.social · 10/07/2026

Academic seminar posters that include a headshot and abstract should really be designed like WANTED ads

2171

Oded Rechavi @odedrechavi.bsky.social · 07/07/2026

I think it's basically your email so we know you're a real person

000

Oded Rechavi @odedrechavi.bsky.social · 06/07/2026

And feel free to run our paper in QED Science, the review will help you understand it better and tell you all about the strengthes and the weaknesses, including what gaps are left to bridge in future studies

110

Oded Rechavi @odedrechavi.bsky.social · 06/07/2026

Check out Ethan's new paper on how the brain helps protect fertility from heat-induced damage, now out in @currentbiology.bsky.social! Ethan will be opening his own lab in London in a few months - I highly recommend joining him (students/post-docs/technicians) www.sciencedirect.com/science/arti...

2164

Oded Rechavi @odedrechavi.bsky.social · 30/06/2026

🥂 Congrats Yehu, super well deserved!

Yehu Moran @yehumoran.bsky.social · 30/06/2026

Reposted by Oded Rechavi

Humbled and incredibly grateful to have been elected to the @embo.org Membership. Many thanks to everyone who has supported me throughout this journey, Mentors, Collaborators and Trainees alike. Special thanks to @odedrechavi.bsky.social for proposing my nomination. It is a tremendous honor. 🙏

10547

Reposted by Oded Rechavi

eric lai @lucksmith.bsky.social · 28/06/2026

Crete Fly 2026 is a wrap! brain overflowing with info, and heart full with joy, after a week of stimulating talks, discussions, and catching up with old and new friends spanning 30 years of my life. also rocked the OAC with Cassandra Extavour for the first time, the fly community is the BEST!! 🪰🇬🇷

35611

Yes, please send it to my gmail and i'll also connect the right people. Thank you!

HI Jon, thank you very much! You are right, we should have QED say more about the strengths as well. Currently you have the Novel aspecet in the Overview section but there should be more.

120

Yes, perhaps we should move to email?

110

Thank you very much, and I also agree about "one score reflecting everything". We'll make the scores more granular (we already have multiple scores behind the scenes. And as I said (and you know), aside from the score there's the detailed review that comes with it.

did you press "generate"? once you do it will appear (in seconds) in the "overview" tab.

100

Anna Poetsch @apoetsch.bsky.social · 26/06/2026

Thanks Marco. And of course I understand. I think it's also a mix reaction to AI in general and clinging to the current system. If we tried to change the system and released something that doesn't upset a few people, we didn't do anything.

100

Reposted by Oded Rechavi

Thank you for doing this! You are tackling a problem without an immediate solution, so things will not be perfect right away, and there will always be criticism. I find the reviews very helpful and also the scoring. Please keep going!

111

We'll do more than that - did you check the "insight feed"? (on the QED platform)

000

Currently there's one score showed, which is a combination of validity and originality. But when you'll see the review you'll see how detailed it is, and also that you get validity assessment for every claim in the paper (per claim not per paper). In the future there'll be more scores.

i meant "run the paper" not run the review... sorry

100

Run the review and you'll see the score, together with the review

100

Sign up, not "sign up", it's free and basically one or two clicks... When you log in you have more than the scores, you can get feedback on manuscript and also get reading recommendations. But Mark, really, why won't you check it before criticizing it?

031

Hi Marco, did you check the "Insight Feed" in QED? I think maybe it's what you're looking for...

100

Scores are an arms race and we need to work at it. The scores are LIVE (updating when knowledge accumulates), they will be more granular in the future (not one score, many) and already now it's not just a score it's a detailed review

we remove the names and affiliations and scored papers that are above the knowledge cutoffs. We analyze the data searching for biases (any that we can think of) and we started publishing all kinds of stats about that.

100