adn 👣
@adnrv.bsky.social
adin meets the blue sky 👁️🤖💻📚
Prof. at IFI UiO DSB
#CV and #ML (mostly)
Prof. at IFI UiO DSB
#CV and #ML (mostly)
Pinned
adn 👣
@adnrv.bsky.social
· Nov 28
"Read papers with the mindset: Why are these lying bastards lying to me?" -- David Forsyth
www.youtube.com/watch?v=9Y7N...
www.youtube.com/watch?v=9Y7N...
Reposted by adn 👣
Call for emergency reviewers #NeurIPS2026
I'm looking for two emergency reviewers:
- Multimodal (text + vision) benchmark
- AI for Science (text-focused) benchmark
If you have experience in either domains, send me a DM including your OpenReview profile and Google Scholar links
I'm looking for two emergency reviewers:
- Multimodal (text + vision) benchmark
- AI for Science (text-focused) benchmark
If you have experience in either domains, send me a DM including your OpenReview profile and Google Scholar links
July 7, 2026 at 3:50 PM
Call for emergency reviewers #NeurIPS2026
I'm looking for two emergency reviewers:
- Multimodal (text + vision) benchmark
- AI for Science (text-focused) benchmark
If you have experience in either domains, send me a DM including your OpenReview profile and Google Scholar links
I'm looking for two emergency reviewers:
- Multimodal (text + vision) benchmark
- AI for Science (text-focused) benchmark
If you have experience in either domains, send me a DM including your OpenReview profile and Google Scholar links
Reposted by adn 👣
CVPR@Paris 2026 🇫🇷 — June 1st, co-organised by ELLIS Unit Paris. A one-day local event ahead of CVPR, open to all. Oral & poster sessions for CVPR 2026, CVPR workshops & ICLR 2026 papers.
🔗 https://cvprinparis.github.io/CVPR2026InParis/
🔗 https://cvprinparis.github.io/CVPR2026InParis/
May 26, 2026 at 12:15 PM
CVPR@Paris 2026 🇫🇷 — June 1st, co-organised by ELLIS Unit Paris. A one-day local event ahead of CVPR, open to all. Oral & poster sessions for CVPR 2026, CVPR workshops & ICLR 2026 papers.
🔗 https://cvprinparis.github.io/CVPR2026InParis/
🔗 https://cvprinparis.github.io/CVPR2026InParis/
Reposted by adn 👣
We are happy to share that 9 tutorials have been accepted for #ECCV2026. The tutorials will take place on the 8th and 9th September. The chosen list can be found here, with more information to be added.
eccv.ecva.net/Conferences/...
eccv.ecva.net/Conferences/...
May 21, 2026 at 1:08 PM
We are happy to share that 9 tutorials have been accepted for #ECCV2026. The tutorials will take place on the 8th and 9th September. The chosen list can be found here, with more information to be added.
eccv.ecva.net/Conferences/...
eccv.ecva.net/Conferences/...
Reposted by adn 👣
Parallels are drawn between Physics and the Theory of Deep Learning, I love it.
arxiv.org/pdf/2604.21691
arxiv.org/pdf/2604.21691
April 29, 2026 at 4:51 PM
Parallels are drawn between Physics and the Theory of Deep Learning, I love it.
arxiv.org/pdf/2604.21691
arxiv.org/pdf/2604.21691
Reposted by adn 👣
Where learning theory meets modern ML:
Learning Theory Summer School (22-25 Jun) + Workshop (26-27 Jun)
📍 Copenhagen 🇩🇰
📚 Generalisation, optimisation, RL & responsible ML
By: @aicentre.dk · Danish Data Science Academy
⏰ Register by 25 May
🔗 All info: https://ltsscopenhagen.github.io/
Learning Theory Summer School (22-25 Jun) + Workshop (26-27 Jun)
📍 Copenhagen 🇩🇰
📚 Generalisation, optimisation, RL & responsible ML
By: @aicentre.dk · Danish Data Science Academy
⏰ Register by 25 May
🔗 All info: https://ltsscopenhagen.github.io/
April 27, 2026 at 12:04 PM
Where learning theory meets modern ML:
Learning Theory Summer School (22-25 Jun) + Workshop (26-27 Jun)
📍 Copenhagen 🇩🇰
📚 Generalisation, optimisation, RL & responsible ML
By: @aicentre.dk · Danish Data Science Academy
⏰ Register by 25 May
🔗 All info: https://ltsscopenhagen.github.io/
Learning Theory Summer School (22-25 Jun) + Workshop (26-27 Jun)
📍 Copenhagen 🇩🇰
📚 Generalisation, optimisation, RL & responsible ML
By: @aicentre.dk · Danish Data Science Academy
⏰ Register by 25 May
🔗 All info: https://ltsscopenhagen.github.io/
Reposted by adn 👣
From the V-Jepa2 presentation:
Action anticipation in EPIC-KITCHENS is "a very challenging task and still there’s head room for improvement on this task".
V-Jepa2 establishes a new SOTA but still <65% on both verb and noun acc.
Mido Assran at the @iclr-conf.bsky.social workshop on World Models
Action anticipation in EPIC-KITCHENS is "a very challenging task and still there’s head room for improvement on this task".
V-Jepa2 establishes a new SOTA but still <65% on both verb and noun acc.
Mido Assran at the @iclr-conf.bsky.social workshop on World Models
April 27, 2026 at 2:35 PM
From the V-Jepa2 presentation:
Action anticipation in EPIC-KITCHENS is "a very challenging task and still there’s head room for improvement on this task".
V-Jepa2 establishes a new SOTA but still <65% on both verb and noun acc.
Mido Assran at the @iclr-conf.bsky.social workshop on World Models
Action anticipation in EPIC-KITCHENS is "a very challenging task and still there’s head room for improvement on this task".
V-Jepa2 establishes a new SOTA but still <65% on both verb and noun acc.
Mido Assran at the @iclr-conf.bsky.social workshop on World Models
Reposted by adn 👣
Software Engineer position open at U. Copenhagen and @aicentre.dk on the Narratives project, co-advised by @ysiglidis.bsky.social, @mariaa.bsky.social (CU Boulder), and me: candidate.hr-manager.net/ApplicationI...
Software Engineer - Department of Computer Science
We are seeking a highly capable and motivated Software Engineer to design and develop a state-of-the-art scalable infrastructure for digital narratives. This wo
candidate.hr-manager.net
April 27, 2026 at 2:50 PM
Software Engineer position open at U. Copenhagen and @aicentre.dk on the Narratives project, co-advised by @ysiglidis.bsky.social, @mariaa.bsky.social (CU Boulder), and me: candidate.hr-manager.net/ApplicationI...
Reposted by adn 👣
Message from #ECCV2026 Program Chairs: We recognise that missing review warning emails may feel excessive, especially for those of you who consistently complete your reviews on time (thank you).
April 19, 2026 at 2:24 PM
Message from #ECCV2026 Program Chairs: We recognise that missing review warning emails may feel excessive, especially for those of you who consistently complete your reviews on time (thank you).
Reposted by adn 👣
What are your all time favorite textbooks? Here are a few of mine.
April 19, 2026 at 6:53 PM
What are your all time favorite textbooks? Here are a few of mine.
Reposted by adn 👣
Looking forward to upgrading all my Ubuntu machines to 26.04 LTS this week
Lots of things to follow the upgrade
Lots of things to follow the upgrade
April 20, 2026 at 4:14 AM
Looking forward to upgrading all my Ubuntu machines to 26.04 LTS this week
Lots of things to follow the upgrade
Lots of things to follow the upgrade
Reposted by adn 👣
In case you haven't seen the mails, we are now a little more than two days from the #ECCV2026 review deadline 😉 Please note the deadline is 11 pm CET, April 21st!
a black and white clock with a spiral in the center
ALT: a black and white clock with a spiral in the center
media.tenor.com
April 19, 2026 at 2:33 PM
In case you haven't seen the mails, we are now a little more than two days from the #ECCV2026 review deadline 😉 Please note the deadline is 11 pm CET, April 21st!
Reposted by adn 👣
April 13, 2026 at 10:17 AM
The job of the program committee of confs is to make a binary decision
The decision should be based on the technical contributions, correctness and soundness of the work
If the work has no more concerns, accept it!! It is binary not fuzzy
-- #ICML weak accept frustration
The decision should be based on the technical contributions, correctness and soundness of the work
If the work has no more concerns, accept it!! It is binary not fuzzy
-- #ICML weak accept frustration
April 8, 2026 at 6:46 AM
The job of the program committee of confs is to make a binary decision
The decision should be based on the technical contributions, correctness and soundness of the work
If the work has no more concerns, accept it!! It is binary not fuzzy
-- #ICML weak accept frustration
The decision should be based on the technical contributions, correctness and soundness of the work
If the work has no more concerns, accept it!! It is binary not fuzzy
-- #ICML weak accept frustration
Reposted by adn 👣
As an AC of #ICML2026, I replied to all AC confidential comments at least for acknowledging their comments because that's how I want an AC to behave rather than keep silence
As an author, I submitted AC confidential comments. There is no response or indication that they read
As an author, I submitted AC confidential comments. There is no response or indication that they read
April 7, 2026 at 12:35 AM
As an AC of #ICML2026, I replied to all AC confidential comments at least for acknowledging their comments because that's how I want an AC to behave rather than keep silence
As an author, I submitted AC confidential comments. There is no response or indication that they read
As an author, I submitted AC confidential comments. There is no response or indication that they read
Reposted by adn 👣
🏹 Job alert: Postdoctoral position in method development for spatio-temporal medical data at the Arctic University of Norway (UiT)
📍 Tromsø 🇳🇴
⏰ Apply by 26 April
🔗 https://bit.ly/4skTQLe
📍 Tromsø 🇳🇴
⏰ Apply by 26 April
🔗 https://bit.ly/4skTQLe
Postdoctoral Research Fellow in Machine Learning (297935) | UiT The Arctic University of Norway
Job title: Postdoctoral Research Fellow in Machine Learning (297935), Employer: UiT The Arctic University of Norway, Deadline: Sunday, April 26, 2026
bit.ly
April 6, 2026 at 6:34 AM
🏹 Job alert: Postdoctoral position in method development for spatio-temporal medical data at the Arctic University of Norway (UiT)
📍 Tromsø 🇳🇴
⏰ Apply by 26 April
🔗 https://bit.ly/4skTQLe
📍 Tromsø 🇳🇴
⏰ Apply by 26 April
🔗 https://bit.ly/4skTQLe
Reposted by adn 👣
👨🔬Welcome to ELLIS Unit Jena, Prof. Dr. Alexander Binder !
Professor for Multimodal Machine Learning and Principal Investigator at @scadsai.bsky.social and @unileipzig.bsky.social focuses on one of the most critical challenges in modern AI: making multimodal models more explainable and trustworthy.
Professor for Multimodal Machine Learning and Principal Investigator at @scadsai.bsky.social and @unileipzig.bsky.social focuses on one of the most critical challenges in modern AI: making multimodal models more explainable and trustworthy.
March 25, 2026 at 1:13 PM
👨🔬Welcome to ELLIS Unit Jena, Prof. Dr. Alexander Binder !
Professor for Multimodal Machine Learning and Principal Investigator at @scadsai.bsky.social and @unileipzig.bsky.social focuses on one of the most critical challenges in modern AI: making multimodal models more explainable and trustworthy.
Professor for Multimodal Machine Learning and Principal Investigator at @scadsai.bsky.social and @unileipzig.bsky.social focuses on one of the most critical challenges in modern AI: making multimodal models more explainable and trustworthy.
Reposted by adn 👣
ICLR 2026: Identity leakage during the review process
ICML 2026: Two-policy framework for "LLM reviewing"
NeurIPS 2026: Sanction list(s?)
2026 must be a tough year for ML conference organizers...
ICML 2026: Two-policy framework for "LLM reviewing"
NeurIPS 2026: Sanction list(s?)
2026 must be a tough year for ML conference organizers...
March 26, 2026 at 12:28 AM
ICLR 2026: Identity leakage during the review process
ICML 2026: Two-policy framework for "LLM reviewing"
NeurIPS 2026: Sanction list(s?)
2026 must be a tough year for ML conference organizers...
ICML 2026: Two-policy framework for "LLM reviewing"
NeurIPS 2026: Sanction list(s?)
2026 must be a tough year for ML conference organizers...
Reposted by adn 👣
Quo vadis deep learning?
March 23, 2026 at 10:43 PM
Quo vadis deep learning?
Reposted by adn 👣
Today NeurIPS is announcing our official satellite event in Paris.
After responding to the call from Ellis following the success of EurIPS in December, we are pleased to reach a new milestone by joining forces with the NeurIPS organizing committee for the 2026 edition.
After responding to the call from Ellis following the success of EurIPS in December, we are pleased to reach a new milestone by joining forces with the NeurIPS organizing committee for the 2026 edition.
March 23, 2026 at 9:17 PM
Today NeurIPS is announcing our official satellite event in Paris.
After responding to the call from Ellis following the success of EurIPS in December, we are pleased to reach a new milestone by joining forces with the NeurIPS organizing committee for the 2026 edition.
After responding to the call from Ellis following the success of EurIPS in December, we are pleased to reach a new milestone by joining forces with the NeurIPS organizing committee for the 2026 edition.
Reposted by adn 👣
💬 AI for Good Webinar Series 'From Molecules to Models' - today with Alexander Binder 🇩🇪 on 'Explainable AI in cases when you cannot spot the dogs or cats'.
📅 Mon, Mar 23
🕓 17:00-18:00 CET
🔗 Registration: https://bit.ly/4sP4URk
ℹ️ More info: https://bit.ly/47Qclim
📅 Mon, Mar 23
🕓 17:00-18:00 CET
🔗 Registration: https://bit.ly/4sP4URk
ℹ️ More info: https://bit.ly/47Qclim
March 23, 2026 at 7:06 AM
💬 AI for Good Webinar Series 'From Molecules to Models' - today with Alexander Binder 🇩🇪 on 'Explainable AI in cases when you cannot spot the dogs or cats'.
📅 Mon, Mar 23
🕓 17:00-18:00 CET
🔗 Registration: https://bit.ly/4sP4URk
ℹ️ More info: https://bit.ly/47Qclim
📅 Mon, Mar 23
🕓 17:00-18:00 CET
🔗 Registration: https://bit.ly/4sP4URk
ℹ️ More info: https://bit.ly/47Qclim
Reposted by adn 👣
"Authors should not use negative v-spaces to change the template layout."
The template layout:
The template layout:
March 6, 2026 at 7:56 AM
"Authors should not use negative v-spaces to change the template layout."
The template layout:
The template layout:
Reposted by adn 👣
TIL that
- "state of the art" is a noun, so we compare to the state of the art
- "state-of-the-art" is an adjective, so we can write about the state-of-the-art model on, say, edge-detection or thresholding or binarization.
- "state of the art" is a noun, so we compare to the state of the art
- "state-of-the-art" is an adjective, so we can write about the state-of-the-art model on, say, edge-detection or thresholding or binarization.
March 4, 2026 at 9:02 PM
TIL that
- "state of the art" is a noun, so we compare to the state of the art
- "state-of-the-art" is an adjective, so we can write about the state-of-the-art model on, say, edge-detection or thresholding or binarization.
- "state of the art" is a noun, so we compare to the state of the art
- "state-of-the-art" is an adjective, so we can write about the state-of-the-art model on, say, edge-detection or thresholding or binarization.
Reposted by adn 👣
The #ECCV2026 paper registration deadline has passed. No new papers may be registered.
The author list is final. No additions allowed; reordering is permitted before submission deadline.
Please do not email us about these matters. The rules are strict, no exceptions.
The author list is final. No additions allowed; reordering is permitted before submission deadline.
Please do not email us about these matters. The rules are strict, no exceptions.
February 27, 2026 at 4:44 PM
The #ECCV2026 paper registration deadline has passed. No new papers may be registered.
The author list is final. No additions allowed; reordering is permitted before submission deadline.
Please do not email us about these matters. The rules are strict, no exceptions.
The author list is final. No additions allowed; reordering is permitted before submission deadline.
Please do not email us about these matters. The rules are strict, no exceptions.
@arxiv.bsky.social can we get compound last names working on the website
?
Two last names get confused with names in the current setup
We need a way to define what is a first name and a last name
-- with regards all latin people
?
Two last names get confused with names in the current setup
We need a way to define what is a first name and a last name
-- with regards all latin people
February 27, 2026 at 4:42 PM
@arxiv.bsky.social can we get compound last names working on the website
?
Two last names get confused with names in the current setup
We need a way to define what is a first name and a last name
-- with regards all latin people
?
Two last names get confused with names in the current setup
We need a way to define what is a first name and a last name
-- with regards all latin people