IAPR TC 2

The IAPR TC2 promotes interaction and collaboration among researchers working on Structural and Syntactical Pattern Recognition.

[IAPR TC 2 @iaprtc2]

Two months left to the S+SSPR 2026 submission deadline 👀 Get your papers ready 📝

[IAPR TC 2 @iaprtc2]

The TC2 is pleased to invite you to submit your work to S+SSPR 2026, which will be held 24–26 August 2026 in Bern, Switzerland. More info on the event can be found at https://t.co/3jQ8h13fEg

[IAPR TC 2 @iaprtc2]

The deadline of S+SSPR 2026 has been extended to May 25, 2026. This deadline is final. More info can be found at https://t.co/3jQ8h13fEg

[IAPR TC 2 @iaprtc2]

In the latest research spotlight on the TC2 website, we interview Prof. Jun Liu, Chair in Digital Health at Lancaster University. Read the full interview at https://t.co/EbQ8XHs4wt

[IAPR TC 2 @iaprtc2]

The TC2 is pleased to invite you to submit your work to S+SSPR 2026, which will be held 24–26 August 2026 in Bern, Switzerland. More info on the event can be found at https://t.co/3jQ8h13fEg

[IAPR TC 2 @iaprtc2]

And just like that the 2nd IAPR Summer School on Deep Learning on Graphs is over. 3 days packed with social events and lectures from researchers working at the interface between graph theory and machine learning. Thanks to all those who participated! 👨‍💻🍷🍝

[IAPR TC 2 @iaprtc2]

The 2nd IAPR Summer School on Deep Learning on Graphs is taking place in Bertinoro, Italy https://t.co/VyRwTLyW8h

[IAPR TC 2 @iaprtc2]

The event will feature a keynote by Prof. Marco Gori (University of Siena) @KleineBottleM and lectures by Prof. Richard Wilson (University of York), Prof. Emanuele Rodolà (Sapienza University) @EmanueleRodola, and Dr. Nils Kriege (University of Vienna) @nmkriege.

[IAPR TC 2 @iaprtc2]

Just over 3 weeks left to apply for #DLG2025, the 2nd IAPR Summer School on Deep Learning on Graphs!

Thanks to the support of the #IAPR we are able to provide a number of scholarships to financially support students in need. More info at https://t.co/LMo5VGivm3

[IAPR TC 2 @iaprtc2]

This month's research spotlight features Prof. Jun Zhou, the current president of the Australian Pattern Recognition Society. https://t.co/3y5FqcUF0N

[IAPR TC 2 @iaprtc2]

The TC2 is glad to announce the introduction of the "research spotlight", a new section of the TC2 website dedicated to showcasing research achievements of members of the TC2 community. Get in touch if you want your research to be featured on our page https://t.co/EbQ8XHrwGV

[IAPR TC 2 @iaprtc2]

We are excited to announce that the 2nd IAPR Summer School on Deep Learning on Graphs will be held in Bertinoro, Italy, on July 7-9 2025. More info at https://t.co/LMo5VGhXwv

[IAPR TC 2 @iaprtc2]

LLMs: modern-day oracles or bullshit machines?
https://t.co/5VvyFOiywa

[Bindu Reddy @bindureddy]

Behold! the power of magic intelligence 😂🍓

[Andrej Karpathy @karpathy]

It's a bit sad and confusing that LLMs ("Large Language Models") have little to do with language; It's just historical. They are highly general purpose technology for statistical modeling of token streams. A better name would be Autoregressive Transformers or something.

They don't care if the tokens happen to represent little text chunks. It could just as well be little image patches, audio chunks, action choices, molecules, or whatever. If you can reduce your problem to that of modeling token streams (for any arbitrary vocabulary of some set of discrete tokens), you can "throw an LLM at it".

Actually, as the LLM stack becomes more and more mature, we may see a convergence of a large number of problems into this modeling paradigm. That is, the problem is fixed at that of "next token prediction" with an LLM, it's just the usage/meaning of the tokens that changes per domain.

If that is the case, it's also possible that deep learning frameworks (e.g. PyTorch and friends) are way too general for what most problems want to look like over time. What's up with thousands of ops and layers that you can reconfigure arbitrarily if 80% of problems just want to use an LLM?

I don't think this is true but I think it's half true.

[IAPR TC 2 @iaprtc2]

Tuesday's keynote: Integrating machine learning into safety-critical systems, by Prof. Dietterich

[IAPR TC 2 @iaprtc2]

Day 2 of S+SSPR 2024 starting now

[IAPR TC 2 @iaprtc2]

Monday's Keynote: Unlocking Neural Composition, by Prof. Rodolà

[IAPR TC 2 @iaprtc2]

S+SSPR 2024 starting now in the beautiful setting of Ca' Dolfin https://t.co/I7kpkxmYrv

[IAPR TC 2 @iaprtc2]

The program of S+SSPR 2024 is online 📋https://t.co/ZrkglrBAuM

19 talks, 2 keynotes
- Emanuele Rodolà (Sapienza University of Rome)
- Thomas Dietterich (Oregon State University, 2024 Pierre Devijver Award winner)

Join us in Venice on 9-10 Sept 🚣🇮🇹

[IAPR TC 2 @iaprtc2]

[S+SSPR 2024 Keynote Speakers]
We are glad to announce that the keynote speakers for S+SSPR 2024 will be Prof. Thomas G. Dietterich (2024 Pierre Devijver Award winner) and Prof. Emanuele Rodolà. More details will be announced in due course on https://t.co/9cBXuJO5D5