Harvard VCG retweeted
We demonstrate that SmartEM achieves up to a ~7-fold acceleration of image acquisition time for connectomic samples using a commercial single-beam SEM in samples from nematodes, mice, and the human brain.
Smarter Microscopes, Faster Maps: Lichtman Lab’s “SmartEM” Brings AI to Connectomics #research
mcb.harvard.edu/department/n… @naturemethods @hseas @MIT @harvardbrainsci @HarvardGSAS @YaronMeirovitch @corefpark @hpfister @VMurthyLab @RachelleGaudet
Harvard VCG retweeted
MIT Professional Education offers a three-day in-person or live online course on June 2-4, 2025, on
AI for Scientific Discovery
professional.mit.edu/course-…
Harvard VCG retweeted
This summer, I’m interning at @E11BIO in SF as an ML fellow! We'll further push my previous work on automated connectome proofreading. Let’s have coffee and chat about AI & vis for bio/neuro if you're in the area! Collab w/ @srinituraga 🧠 #connectomics #AI @Convergent_FROs
Harvard VCG retweeted
If you're an undergrad with strong programming skills, and would like to work on LLMs + Human Memory at @Harvard, drop me a message.
Harvard VCG retweeted
📢 Excited to share our #NeurIPS 2024 paper SocialGPT, a modular framework with greedy segment prompt optimization for social relation reasoning.
🚀🚀🚀
Paper: arxiv.org/pdf/2410.21411
Code: github.com/Mengzibin/SocialG…
Website: mengzibin.github.io/SocialGP…
Harvard VCG retweeted
Interested in attending the 2024 Boston VIS Summer School on August 13-14? Sign up for the mailing list: docs.google.com/forms/d/e/1F…
#BostonMA #VIS #ieeevis #acmchi #DataVisualization #NewEngland @Northeastern @UMass @mitvis @HarvardVCG @WPI @BrownUniversity @Harvard @MIT @BUCSC
Harvard VCG retweeted
Thanks @scicoalition for hosting Harvard at “The Science of Sports” event, with support from our very own Congresswoman @RepLoriTrahan !
Had a blast sharing our data vis research @sportsbuddyai for sports storytelling on behalf of @HarvardVCG to a diverse group of sports lovers🎉
Harvard VCG retweeted
Phinally Done!✅ A picture is worth a thousand words, and a PhD journey is worth countless thank-yous.
I’m deeply grateful to my advisor @hpfister, mentors, family and friends for their love & support, leading me to go this far and be ready for next chapter!💪#harvardclassof2024
Harvard VCG retweeted
🔬🐝Excited to share our latest paper published
@IEEE_TMI: A novel benchmark for domain adaptive synapse detection, consisting of 14 electron microscopy image volumes from a biologically diverse set of Megaphragma viggianii brain regions.
Early Access Papers Just Posted: WASPSYN: A Challenge for Domain Adaptive Synapse Detection in Microwasp Brain Connectomes. Read the paper: ieeexplore.ieee.org/document…
Harvard VCG retweeted
Very nice. Congrats @stardazed0 and team 👏
This year is the 10th anniversary of the Google Research’s Connectomics team! In celebration, today with @MCB_Harvard we’re publishing a 1.4 petabyte human brain connectome with 57k cells and 150M synapses in @ScienceMagazine → goo.gle/4a8SpWT & sites.research.google/neural…
Harvard VCG retweeted
I am feeling very happy that I could contribute to this massive research study. We added community-based proofreading with CAVE to this dataset. Everyone can edit/update the AI-generated automated segmentation. ➡️bit.ly/4dAPxoB @sdorkenw @fcollman JinhanChoi
This year is the 10th anniversary of the Google Research’s Connectomics team! In celebration, today with @MCB_Harvard we’re publishing a 1.4 petabyte human brain connectome with 57k cells and 150M synapses in @ScienceMagazine → goo.gle/4a8SpWT & sites.research.google/neural…
Harvard VCG retweeted
What if we could decode the mysteries of how we think? 🧐 New research from our Connectomics team and @Harvard University reveals surprising new insights about our human brain, using scientific imaging and our advanced AI algorithms. Learn more → goo.gle/44zxGdP
Harvard VCG retweeted
Let’s envision the future of sports and personal analytics with XR visualizations! Anyone interested in XR and vis is encouraged to submit to our workshop held this year at VIS!
This Post is from an account that no longer exists. Learn more
Harvard VCG retweeted
Lichtman Lab Teams With Google to Map 150 Million Synapses in Human Brain Sample @alex_shapsoncoe @mjanuszewski @N8293Neha @jakobtroidl @SjostedtEvelina @hpfister @hseas @Harvard @ScienceMagazine @Google mcb.harvard.edu/department/n…
Harvard VCG retweeted
All my journey in @HarvardVCG has come to an end. Have an unforgettable and dreamy experience these 3 months. Learn a lot and grow up so much from both life and work. Miss u guys and hope we will have chance to meet again. Thank u for having me.❤️
Harvard VCG retweeted
Felt grateful to be back to where I grew up and share my journey breaking boundaries at Tainan Girls High School & Tainan First High School! The overwhelming engagement and curiosity have amazed me. Knowledge sharing is key to inclusion and innovation!#WomenInSTEM #Diversity
Harvard VCG retweeted
📣 Our interactive motif search tool is already available for this new optic lobe connectome. Just sketch the motif that you are interested in, and voilà!🖱️neuprint.janelia.org/?datase…
🧠Janelia scientists have reached another milestone in connectomics: A wiring diagram of the fruit fly visual system.
With 50K+ neurons, the optic lobe connectome provides the fullest picture yet of one of the most important parts of the nervous system.
janelia.org/news/seeing-is-b…
Harvard VCG retweeted
Very excited to attend my very first PacificVis at the beautiful campus at Keio University, the oldest university in Japan. Enlightening keynote by Prof. Huamin Qu on data storytelling, where vis research can make broad impact and unleash human creativity
#pacificvis2024
Harvard VCG retweeted
Pardon my self-promotion! In my recent work, I measured how chatbots can drift away from its system prompts within 8 rounds of discourse. Then I introduced a training-free hack in the attention heads to help the model focus on privileged/system prompts. teapaw.org/ke_li_2021/status/1764…
Introducing the Instruction Hierarchy, our latest safety research to advance robustness for prompt injections and other ways of tricking LLMs into executing unsafe actions. More details: arxiv.org/abs/2404.13208











































