@1uc4r0 @dcucomputing Congratulations @1uc4r0 , I am grateful to @dcucomputing for sponsoring my PhD 20 years ago. I had great time and memories there. Best of luck 🤞 in your new chapter
English
George Awad
193 posts
Tomorrow, I'll be joining @dcucomputing as an assistant professor. Looking forward to seeing what the future holds...
English
After five years @uzh_ifi, it was time to say goodbye. Thank you so much to @avibern, @csarasuagar, @oana_inel, and everybody at @DDIS_UZH for our time together, I will always look back to it fondly.
English
George Awad retweetledi
🆕 ACM Multimedia 2024 will be in Melbourne, Australia on 28 October - 1 November 2024 🇦🇺
Follow @ACMMM24 🥳
- Grand Challenge Proposal Submission: Feb 19, 2024
- Workshop Proposal Submission: March 4, 2024
- Regular Papers Submission: April 12, 2024
🔗 2024.acmmm.org
English
Announcing GPT-4, a large multimodal model, with our best-ever results on capabilities and alignment: openai.com/product/gpt-4
English
George Awad retweetledi
On Tues., 12/6, #LIS Associate Professor Vivek Singh (@vivks) is giving a keynote talk, “Toward Fairer Multimodal Information Systems,” @NIST’s TRECVID2022, a workshop that encourages research in content-based video retrieval.
#RUSCIscholarship #RUSCIproud
English
George Awad retweetledi
Renmin University of China at TRECVID 2022: Improving Video Search by
Feature Fusion and Negation Understanding arxiv.org/abs/2211.15039…
TRECVID 2022 Ad-hoc Video Search (AVS) の実験をまとめます。私たちのソリューションは、2 つの新しい技術、すなわち、さまざまな視覚的
日本語
George Awad retweetledi
English
George Awad retweetledi
I am looking to hire one or two undergrad interns on the NIST Pathways program. One is for a sysadmin coder interested in language technology. The other is for a computer scientist interested in growing into IR and NLP.
nist.gov/itl/how-work-u…
English
@ian_soboroff I usually know these things either from nextdoor OR the ring app
English
Is there an app that tells me why that helicopter is hovering over my house for the last half hour?
English
George Awad retweetledi
We are still looking for PhD Students to join our team in Zurich!
#PhDposition #AcademicChatter #AcademicTwitter #PhDCandidate #AcademicPosition #computerscience
English
@twelve_labs @IndexVentures @_jae_lee Congratulations, you should probably join the #TRECVID annual evaluation to test your algorithms
English
Twelve Labs is building the world's most powerful video understanding infrastructure! Thrilled to announce our $5M seed round led by @IndexVentures. Here's our CEO @_jae_lee on how and why Twelve Labs is making search within videos a reality.
link.medium.com/PyTVejxGrob
English
George Awad retweetledi
"Vision Models Are More Robust And Fair When Pretrained On Uncurated Images Without Supervision"
from FAIR Paris & NYC
Large-scale experiment with 10 billion param RegNet pre-trained by SSL with SwAV on 1 billion random public Instagram photos.
arxiv.org/abs/2202.08360
English
George Awad retweetledi
If you are working on video search, understanding, captioning, detection, summarization, or indexing consider joining TRECVID 2022 www-nlpir.nist.gov/projects/tv202…
#Artificial_Intelligence #Video #multimedia @ian_soboroff @asadanwar @gquenot @asmeaton @DocXavi @dimadamen @cgmsnoek @cwngo
English
George Awad retweetledi
ConvNeXt: the debate heats up between ConvNets and Transformers for vision!
Very nice work from FAIR+BAIR colleagues showing that with the right combination of methods, ConvNets are better than Transformers for vision.
87.1% top-1 ImageNet-1k
arxiv.org/abs/2201.03545
1/N
English
George Awad retweetledi
We are looking for a postdoc to work with Ellen Voorhees on test collection building methods. Ideally a candidate would be ready to start in Jan or Feb. US citizenship not required, but you will need to come to the US. Drop me a PM or email if interested.
English
@DocXavi @trecvid_nist Our pleasure. That was a great informative talk. It's always encouraging to see what previous #trecvid team members are working on now
English
Thanks to George Awad (@trecvid) for inviting me as keynote speaker at the #TRECVID Workshop 2021 (@trecvid_nist).
I reviewed our adventures with deep learning from the TRECVID instance task 2014 to our current focus on sign language videos.
imatge.upc.edu/web/publicatio…
GIF
English
George Awad retweetledi
Very insightful thread on the current trend.
Andrej Karpathy@karpathy
The ongoing consolidation in AI is incredible. Thread: ➡️ When I started ~decade ago vision, speech, natural language, reinforcement learning, etc. were completely separate; You couldn't read papers across areas - the approaches were completely different, often not even ML based.
English
George Awad retweetledi
TRECVID is this week. Some nice features of their search task are having a progress set of queries, and using unique relevant as a metric (“novelty task”).
English
Honored to be among them
Swiss National Science Foundation@snsf_ch
#Ambizione: the SNSF has awarded 84 grants to young researchers. They will receive an average of 766,000 francs to run their first independent research project at a higher education institution in Switzerland. #SNSF_Research snf.ch/en/5lCGf3Sam9G…
English
George Awad retweetledi
The datasets were not built to be solved. They were built as tools to understand the problem and the systems we build to “solve” them.
English