Robert Pless
@pless.bsky.social
1.2K followers 300 following 23 posts
Professor at GWU, Developer of @traffickcam and @projectRephoto. Computer vision/AI researcher with mercenary interest in glitter and contemporary art. http://www2.seas.gwu.edu/~pless
Posts Media Videos Starter Packs
pless.bsky.social
I'm excited to share that GWU is hosting a World Bank Symposium on AI & the Future of Human Capital.

For those not steeped in the development, Human Capital == people’s skills, knowledge, and health

www.worldbank.org/en/events/20...

Some flexibility in the deadline is likely to be possible.
AI & the Future of Human Capital in the Global South: A World Bank-George Washington University Knowledge Symposium
How AI-driven innovations can help address critical human development challenges in low- and middle-income countries. Call for research papers: Submissions due July 18
www.worldbank.org
pless.bsky.social
Finally, papers are routed to specific Area Chairs based on expertise and the research community isn't so big once you are in a specific area, so I'll share that I notice both the community members that put extra effort into writing reviews, and those that don't. Thanks you to those that do! (4/4)
pless.bsky.social
Adding this step, where the executive decision maker has the obligation to object to or approve of them, helps align the authority to make the decisions with the responsibility to get good reviews.
Additionally, the Area Chairs see the name of the reviewer (I'm looking at you, reviewer #2). (3/4)
pless.bsky.social
In the past, as an area chair I would end up with a collection of reviews, some of which sucked, and would somewhat throw up my hands and say "these reviews aren't great and I have to decide based on them, but what can I do ... there are too many papers to do anything else". (2/4)
pless.bsky.social
I'm an Area Chair for NeurIPS this year. I spent most of today reviewing all the reviews for my 13 assigned papers to highlight reviews that were insufficient. There were a few reviews that were incredibly shallow and some had hallmarks of GPT, and this was a chance to reject those. (1/4)
pless.bsky.social
Question: Does anyone have a pre-computed faiss indices for large datasets (LAION 2B, 5B or similar) that have been used to train CLIP-like models?

The one below doesn't seem to work anymore
knn5.laion.ai
pless.bsky.social
I’m curious of any of my friends are at the AI expo in DC today? I’ll be hanging around the George Washington University Trustworthy AI booth for some of the day. Happy to have coffee with anyone!
pless.bsky.social
A view of many kites flapping in the wind at the DC kite festival
pless.bsky.social
The kite festival is one of the best days in DC
pless.bsky.social
I'm lucky to get to work on important problems with some great researchers; @astylianou.bsky.social is giving a talk about our work on image search for sex trafficking investigations tomorrow (Wednesday) at noon (link below)
Reposted by Robert Pless
csprofkgd.bsky.social
The International Conference on Computational Imaging (ICCP) is coming to my hometown, #Toronto! Excited to serve as a social media co-chair for such a fantastic event.

With an incredible organizing committee and an amazing city to host, how could I say no?
pless.bsky.social
(Click to enlarge, the rainbow is outside of the narrow default thumbnail)
pless.bsky.social
one of the most insane webcam rainbow images I've ever seen, from (volcview.wr.usgs.gov/ashcam-api/i..., but that might be different now!)
pless.bsky.social
Welcome! Lots of great people here!
pless.bsky.social
Birds (and bugs!) flying over DC last night
pless.bsky.social
Cool talk at the AI Aspirations event at the old Newseum
pless.bsky.social
A lazy afternoon in Washington DC, in the style of Xavi Bou
An "ornithography" style composite picture showing trajectories of birds flying in the sky.
pless.bsky.social
My favorite tree in the field by the elementary school close to my house. #gloverPark
pless.bsky.social
The prompt was not “a bird on a floating branch”
pless.bsky.social
I'm excited to be heading today to NeurIPS. If you're there and want to chat, drop me a note!

If you're interested in faculty openings at GWU this year, also drop me a note!
pless.bsky.social
Real estate options are limited in the DC area
pless.bsky.social
TIL that NOAA operates hundreds of buoycams that constantly take panoramic images.
Reposted by Robert Pless
meganakpeters.bsky.social
what happens to retracted papers in regards to LLM training corpuses? do they get ignored, or do gpt4 et al treat them like any other "trustworthy" high profile CNS publication? 🤔

anybody know?
davidho.bsky.social
“Nature, one of the most prestigious journals in scientific publishing, on Tuesday retracted a high-profile paper it had published in March that claimed the discovery of a superconductor that worked at everyday temperatures.”
Room-Temperature Superconductor Discovery Is Retracted
It was the second paper led by Ranga P. Dias, a researcher at the University of Rochester, that the journal Nature has retracted.
www.nytimes.com
pless.bsky.social
All creatures enjoy sunsets?
pless.bsky.social
My friends do cool things!