iiinova — where your digital self keeps evolving

Nanne van Noord @nanne.bsky.social · 13/09/2026

Happy to take a couple: ~Nanne_Van_Noord1

Nanne van Noord @nanne.bsky.social · 22/06/2026

The deadline for VISART has been extended to June 29th, do consider submitting your work on visual analysis of art and culture! #ECCV2026 #ComputerVision

010

Nanne van Noord @nanne.bsky.social · 19/06/2026

Seems the be recurring advice to go see other places (Copenhagen, Lund, ..) rather than Malmö itself 😆

130

Nanne van Noord @nanne.bsky.social · 02/06/2026

This year we have expanded the focus to include both art and culture, in recognition of the impact and potential of Computer Vision on/for our visual culture. I'm very much looking forward to the many inspiring talks and discussions!

001

Nanne van Noord @nanne.bsky.social · 02/06/2026

Exciting news - this year the Vision for Art and Culture (VISART) workshop is taking place at ECCV in Malmö (Sweden) on 8/9 September! Submission deadline: 24th June, 2026 Notification of acceptance: 5th August, 2026 visarts.eu #ECCV2026 #ComputationalHumanities #CulturalAnalytics #ComputerVision

visarts.eu

VISART 2026

110

Reposted by Nanne van Noord

Erik Bekkers @erikjbekkers.bsky.social · 15/04/2026

We're looking for a new colleague at @amlab.bsky.social: Assistant Professor in AI for Science 🔬🤖 World-class ML research, Amsterdam's thriving AI ecosystem (ELLIS, startups, big tech), and some of the best academic labor conditions in Europe ❤️ Deadline: May 30 👉 werkenbij.uva.nl/en/vacancies...

werkenbij.uva.nl

Vacancy — Assistant Professor in AI for Science (AI4Science)

Are you passionate about advancing Machine Learning by integrating insights from the natural sciences? Are you eager to bridge the 3rd (computational) and 4th (<span...

03316

Nanne van Noord @nanne.bsky.social · 19/02/2026

Interested to find out why there are fewer women artists in Dutch museums? Few days left to apply!

000

Nanne van Noord @nanne.bsky.social · 29/01/2026

Congrats @mila-oiva.bsky.social ! Excited to see all the amazing things you'll be doing at FAU 🤩

010

Nanne van Noord @nanne.bsky.social · 19/11/2025

Two students of our lab are presenting an artwork at NeurIPS, how amazing is that? Really impressed with the project openreview.net/pdf?id=BZjSU..., and the video they made for it! www.youtube.com/watch?v=L631...

youtube.com

JUTTERS

YouTube video by Meike

030

Nanne van Noord @nanne.bsky.social · 19/11/2025

Wait, you get summaries of your own papers? That seems like step up from the "I see you work on <insert topic I've not touched in my life>" emails at least

020

Reposted by Nanne van Noord

NeurIPS Europe @neuripseurope.bsky.social · 19/09/2025

And lastly, if @neuripsconf.bsky.social would choose to reverse the decisions on the papers affected by space constraints, we would be happy and able to accommodate their presentation

02611

Nanne van Noord @nanne.bsky.social · 27/07/2025

You're arguing in bad faith, so this will be my last reply. But yes, if you actually want to learn about multimodality then you shouldnt read about MLLM.

100

Nanne van Noord @nanne.bsky.social · 27/07/2025

I'm not sure what the point here is, but if you're going to believe Gemini over actual research done by AI researchers there isn't much more to discuss. If you're willing to actually learn about this then you can start here: arxiv.org/abs/2505.19614, or even here: academic.oup.com/dsh/article/...

100

Nanne van Noord @nanne.bsky.social · 27/07/2025

That's a bit sealion-y, but I'll bite - *artificial* neural networks are a poorly analogy. Those different details also matter a lot; especially because the brain isn't just floating in a jar, it's part of an embodied system.

110

Nanne van Noord @nanne.bsky.social · 27/07/2025

This is where your misunderstanding is happening, as they are not elementary pieces. For the visual tokens a lot of the semantics have already been determined, and hence the interpretations it can arrive at are limited. Brain analogy really doesnt hold here. NN != Brains.

100

Nanne van Noord @nanne.bsky.social · 27/07/2025

Its clearly not; neural nets are a poor analogy for the brain, and clearly don't work the same way.

100

Nanne van Noord @nanne.bsky.social · 27/07/2025

This, plus the (initial) interpretation of the modalities should not be independent - even at the pixel/word-level we may want to interpret differently depending on the other modalities (e.g., sense disambiguation) Partial Information Decomposition has been used to formalise some of this

110

Nanne van Noord @nanne.bsky.social · 27/07/2025

No.. that's not how any of that works 😵‍💫

100

Nanne van Noord @nanne.bsky.social · 27/07/2025

It means I said 'mix' to explain the process, but I obviously know this involves attention - so the Gemini explanation is not meaningfully different. Potential limited: if key visual info is missing, then attention wont recover that. So alot of 'decisions' about visual are made before fusion

100

Nanne van Noord @nanne.bsky.social · 26/07/2025

Ah, I see how you and Gemini misunderstood. I was talking about extracting visual tokens, and mix referred to attention. That doesnt make it meaningfully multimodal; potential of visual tokens is still limited by visual encoder. Anyway, if I wanted to talk to an LLM I would do that directly

110

Nanne van Noord @nanne.bsky.social · 26/07/2025

Please do explain then how whatever you're referring to is different and actually meaningfully multimodal.

100

Nanne van Noord @nanne.bsky.social · 26/07/2025

*all semantic information* is quite the claim; in our experiments they miss a lot of semantics from visual 'text space' in that after the image encoder the visual information is fixed, and mixed with text tokens for seq2text - which is not how multimodality works..

111

Nanne van Noord @nanne.bsky.social · 26/07/2025

Natively is a bit of an exaggeration, as it's mostly just other modalities mapped to text space as input - but this makes their 'understanding' rather shallow

131

Nanne van Noord @nanne.bsky.social · 25/07/2025

This paper on identifying prompted artist names from generated images is such a fun and creative take on data attribution arxiv.org/abs/2507.18633 Wonder if it would do something meaningful for analysing artistic influence for human-made art 🤔

arxiv.org

Identifying Prompted Artist Names from Generated Images

A common and controversial use of text-to-image models is to generate pictures by explicitly naming artists, such as "in the style of Greg Rutkowski". We introduce a benchmark for prompted-artist reco...

051

Nanne van Noord @nanne.bsky.social · 24/07/2025

This paper is 💯 Generally, I have the impression NLP does better at this than CV - but clearly both fields should push studying culture beyond just looking at national identities

050

Nanne van Noord @nanne.bsky.social · 18/07/2025

If the priority is to dunk on people that know less about AI, instead of being accurate, that could be a conclusion I guess.

000

Nanne van Noord @nanne.bsky.social · 18/07/2025

It would be weird to describe this 2012 system, that is doing search, as an SVM classifier doing search: www.robots.ox.ac.uk/%7Evgg/publi... Similarly, I wouldn't describe an LLM that translates a query to a destination for a Waymo as an 'LLM driving a car'

robots.ox.ac.uk

Visual Geometry Group - University of Oxford

Computer Vision group from the University of Oxford

100

Nanne van Noord @nanne.bsky.social · 18/07/2025

I'm not questioning your definition of searching, I'm questioning your use of "LLMs". I don't think defining an LLM as a transformer-based NN is inaccurate, in which case it isn't doing search by itself, and then it would be fine to argue that it can only hallucinate.

200

Nanne van Noord @nanne.bsky.social · 18/07/2025

That statement mostly seems to apply to hosted commercial systems. It takes more than just downloading an LLM from huggingface to have a system that does this. Sure an LLM can be trained to formulate queries and process results, but the system doing the searching is more than 'just' an LLM.

010

Nanne van Noord @nanne.bsky.social · 18/07/2025

Fair, but still meaningful to make the distinction between LLMs and reasoning models, as not all LLMs are reasoning models. Especially if the point is to communicate across silos.

100

Nanne van Noord @nanne.bsky.social · 18/07/2025

Do LLMs do search? Afaik there have been systems built around LLMs that do search, and then send these results back to them (i.e., RAG-like) - but that isn't the same as an LLM doing search.

100

Nanne van Noord @nanne.bsky.social · 17/07/2025

I couldnt find EurIPS registration costs; hopefully they can address this by lowering costs for authors But yes - this has been absurd; especially for those with visa issues - and I do think for that group this is a (minor) improvement

010

Nanne van Noord @nanne.bsky.social · 17/07/2025

Not my intention to defend the requirement for a full registration, but this has been common practice for a while across multiple conferences. The main change of new locations seems primarily that those with US visa issues will be able to present somewhere. But it doesnt really change costs

110

Nanne van Noord @nanne.bsky.social · 17/07/2025

This considers registration only, no? One could register for in person, but not go - folks with visa issues have had to do this

110

Nanne van Noord @nanne.bsky.social · 30/06/2025

This distinction is also useful because it makes it harder to avoid responsibility, as its easy to avoid directly working on surveillance - yet harder to avoid doing CV work that is surveillance-enabling. Unless your position is that these are the same?

000

Nanne van Noord @nanne.bsky.social · 30/06/2025

It's clear that CV can be used for surveillance, however, it still seems meaningful to distinguish between surveillance technology and surveillance-enabling technology. As I would consider 'scene understanding' the latter (at a stretch) but not the former.

120

Nanne van Noord @nanne.bsky.social · 30/06/2025

Going from this to 'human data extraction' seems to involve a leap, where the centrality of personal data seems to disappear. E.g., describing labeling a set of pixels as 'person' as human data extraction seems fair, but calling it surveillance without there being personal data seems like a leap.

120

Nanne van Noord @nanne.bsky.social · 30/06/2025

The paper referenced for the definition (by G.T. Marx) states "A central feature is gathering some form of data connectable to individuals (whether as uniquely identified or as a member of a category).", i.e., gathering of personal data.

110

Nanne van Noord @nanne.bsky.social · 19/06/2025

Consider submitting to our ACM Multimedia workshop on Multimedia Analytics with Multimodal Large Language Models ma-llm25.github.io We've extended the deadline till June 27th, to give you extra chance to submit and join us in discussing this exciting and emerging research area!

ma-llm25.github.io

MA-LLM25 - ACM MM 2025 Workshop

000

Reposted by Nanne van Noord

Hannes Cools @hannescools.bsky.social · 21/05/2025

📣 Call for papers - 𝐆𝐞𝐧𝐀𝐈 & 𝐂𝐫𝐞𝐚𝐭𝐢𝐯𝐞 𝐏𝐫𝐚𝐜𝐭𝐢𝐜𝐞𝐬 📣 Working on topics like creative practices & work, governance, regulation, responsibility in relation to GenAI? Consider submitting your work. All details below 👇 📅 On 17 & 18 December 2025 📍 Amsterdam, NL Please share 🙏 rdt.uva.nl/genai--creat...

rdt.uva.nl

Call for Papers: GenAI & Creative Practices - Responsible Digital Transformations

Conference at the University of Amsterdam, 17 & 18 December 2025

121

Nanne van Noord @nanne.bsky.social · 13/05/2025

The PhD vacancy closes at the end of this week, and the postdoc vacancy next week - if you're interested or have questions don't hesitate to contact me!

011

Nanne van Noord @nanne.bsky.social · 30/04/2025

I'm hiring a PhD and a Postdoc! 📣 The PhD is on learning from feedback for bias analysis in cultural collections, more info at: werkenbij.uva.nl/en/vacancies... The postdoc is on gender representation in generative AI and artistic potentials, more info at: werkenbij.uva.nl/en/vacancies...

0156

Nanne van Noord @nanne.bsky.social · 01/04/2025

Are there textual captions/descriptions of these images? Would be interesting to try out our ground model (arxiv.org/abs/2410.12369) on it!

110

Reposted by Nanne van Noord

UvA Faculty of Humanities @uvahumanities.bsky.social · 19/03/2025

We zien steeds meer door AI gegenereerd beeld om ons heen, maar hoe bevooroordeeld zijn de achterliggende AI-systemen? @nanne.bsky.social, @melvinwevers.bsky.social en Eftychia Stamkou gaan dat samen met partners uit de culturele sector onderzoeken: www.uva.nl/shared-conte...

uva.nl

Hoe beïnvloedt AI de beelden die het genereert?

Nu iedereen met behulp van AI afbeeldingen kan creëren, zien we steeds meer door AI gegenereerd beeld om ons heen. Tegelijkertijd is er maar weinig bekend over hoe de achterliggende AI-systemen werken...

052

Reposted by Nanne van Noord

Cees Snoek @cgmsnoek.bsky.social · 03/02/2025

🤖 TULIP: Token-length Upgraded CLIP by Ivona Najdenkoska, Mohammad Mahdi Derakhshani, Yuki Asano, Nanne van Noord, Marcel Worring, Cees Snoek. 13/n

121

Reposted by Nanne van Noord

Neil Cohn @neilcohn.bsky.social · 14/12/2024

I’m excited to share my new paper with @nanne.bsky.social analyzing the style of comics in our TINTIN Corpus (1030 comics from 144 countries). We characterized the sequentiality of panels using their similarity, computed using similar tasks as in language modeling bmvc2024.org/proceedings/...

Characterization of modeling task for comics sequencing

1227

Reposted by Nanne van Noord

David Picard @davidpicard.eurosky.social · 10/12/2024

😍I'm super happy of how this paper turned out! Pretty cool use of diffusion IMHO. Check the demo, upload an image and get its location: huggingface.co/spaces/nicol...

huggingface.co

Plonk - a Hugging Face Space by nicolas-dufour

Around the World in 80 Timesteps

3407

Reposted by Nanne van Noord

Noa Garcia @noagarciad.bsky.social · 10/12/2024

What better time to announce a new paper than during NeurIPS and ACCV? happy happy happy to introduce NADA, our latest work on object detection in art! 🎨 with amazing collaborators: @patrick-ramos.bsky.social, @nicaogr.bsky.social, Selina Khan, Yuta Nakashima

1226