IDEAS NCBR

Presentation of publication titled Knowledge-Driven Bayesian Uncertainty Quantification for Reliable Fake News Detection by Julia Puczyńska is being held now at the 28th European Conference on Artificial Intelligence (#ECAI2025) at the Nettuno Room.

#ECAI2025 #IDEASNCBR #AI #FakeNewsDetection

Dear #ECAI2025 Attendees,

Tomorrow, during #ECAI 2025 our colleague Julia Puczyńska will present publication: Knowledge-Driven Bayesian Uncertainty Quantification for Reliable Fake News Detection.

Come and listen about fake detection. See you tomorrow 🤝

#ECAI2025 #AI #FakeNewsDetection

Our research presented at ICCV 2025! 🌴

Maksymilian Kulicki presented "AutoSeg: Auto-Vocabulary Segmentation" at the International Conference on Computer Vision (#ICCV), held October 19-25 in Honolulu, Hawaii.

👉 Project page: ozzyou.github.io/autoseg.gith...

#ICCV2025 #AI #ComputerVision #Research

Just 1️⃣ day left!

During 28th European Conference on Artificial Intelligence our colleague Julia Puczyńska will present work titled: Knowledge-Driven Bayesian Uncertainty Quantification for Reliable Fake News Detection.

See you tomorrow at #ECAI2025 in Bologna🤝

#ECAI #AI #IDEASNCBR #ECAI2025

Today during the #MLinPL poster session Michał Krutul presented work: Projected Compression: Trainable Projection for Efficient Transformer Compression.
(arxiv.org/abs/2506.22255).

The paper has been accepted to #NeurIPS 2025!

#NeurIPS #IDEASNCBR #MLinPL

Program of ML in PL Conference 2025 ML in PL Conference 2025 is an event focused on the best of Machine Learning both in academia and in business.

Today Dominik Jacek Bogucki, Maciej Pióro & Grzegorz Rypeść their publications during the ongling #MLinPL poster sessions.

That's not all! Michał Krutul will also present his poster. Join tomorrow's poster session 🤝!
📅 October, 17
🕰️ 12:15

conference.mlinpl.org/program#post...

See you tomorrow 😊

Today Mateusz Olko and Mateusz Gajewski delivered a talk about Limits in Causal Discovery and the Path Forward during
#MLinPL.
That's not all about #CasualDiscovery! There will be a tutorial conducted by M.Gajewski and M.Olko about Modern Causal Discovery! 🗓️October, 18 🕑2:00 PM

#MLinPL #IDEASNCBR

Program of ML in PL Conference 2025 ML in PL Conference 2025 is an event focused on the best of Machine Learning both in academia and in business.

Today Dominik Jacek Bogucki, Maciej Pióro & Grzegorz Rypeść their publications during the ongling #MLinPL poster sessions.

That's not all! Michał Krutul will also present his poster. Join tomorrow's poster session 🤝!
📅 October, 17
🕰️ 12:15

conference.mlinpl.org/program#post...

See you tomorrow 😊

#MLinPL2025

Would you like to listen about our scientsts' works? Join Mateusz Olko's and Mateusz Gajewski's talk about Limits in Causal Discovery and the Path Forward.
📅 October, 17
🕤 9:30
📍 Hall B (CfC Session 8)

#MLinPL2025 #IDEASNCBR

Accurate and Efficient Low-Rank Model Merging in Core Space In this paper, we address the challenges associated with merging low-rank adaptations of large neural networks. With the rise of parameter-efficient adaptation techniques, such as Low-Rank Adaptation ...

Daniel Marczak's publication "Accurate and Efficient Low-Rank Model Merging in Core Space" has been accepted to NeurIPS 2025! Congratulations 👏

More about the publication 👇
arxiv.org/abs/2509.17786

Alicja Ziarko's publication has been accepted to #NeurIPS2025‼️

More about the work👉 princeton-rl.github.io/CRTR/
The code to reproduce the experiments is available here👉
github.com/Princeton-RL...

Our colleague Adam Pardyl's publication has been accepted to #NeurIPS2025‼️

The title of the work:FlySearch: Exploring how vision-language models explore
More👉https://arxiv.org/abs/2506.02896

ELIAS & Bosch are inviting to a challenge.

More details: elias-ai.eu/opportunitie...

#AI #SmartBuildings #HVAC #ELIAS #Bosch #MachineLearning #EnergyEfficency #AIChallenge #AICompetition #SustainableBuildings

Daniel Marczak presented his work No Task Left Behind: Isotropic Model Merging with Common and Task-Specific Subspaces" which during Ellis Doctoral Sypmposium (#EDS2025).

More about the publication: arxiv.org/pdf/2502.04959

#IDEASNCBR #AI #EDS2025

Daman Sójka presented his work Adaptive Monocular Depth Estimation with Masked Image Consistency during Ellis Doctoral Symposium (#EDS2025).

Link to the publication: openreview.net/forum?id=qqT...

#IDEASNCBR #AI

Today Adam Pardyl presented his poster FlySearch: Exploring how vision-language explore during Ellis Doctoral Symposium.

The publication introduces FlySearch, a 3D, outdoor, photorealistic environment for searching and navigating to objects in complex scenes.

#AI #EDS2025 #IDEASNCBR

Are you a Master's or PhD student? Willing to be a Start-up Leader? Apply for the summer school How to be a (Startup-) CTO. There are a couple of spots available for ELIAS-affiliated participants. Apart from technical skills, you will have an occasion to gain entrepreneurial & leadership skills.

Our PhD student Maksymilian Kulicki participated in the Interdisciplinary Summer School on Precision Forestry Using LiDAR Technology.

#PrecisionForestry #AI #IDEASNCBR #LiDAR

Security Research Event (SRE 2025) was an event arranged by the European Comission on June 24-25, in Warsaw. We also took part in the event. The projects presented were excellent examples of initiatives that truly contribute to making Europe safer.

#SRE2025 #IDEASNCBR #Security #Innovations #AI

On June 11–12, 2025, we had the honor to participate in Poland Security Congress — an event that brought together representatives from government, business, academic, and the security sector to discuss the challenges Poland faces.

Photos: #KongresBezpieczenstwoPolski #MMCPolska personal materials

It is icml time. Are there folks here doing the same as me, i.e. writing papers?

#icml #icml2025

Michał Tomza: Paper mills to psucie nauki na poziomie systemowym - Forum Akademickie Najbardziej smuci mnie fakt, że to sami badacze pozwalają sobie na nieetyczne zachowania. Jak na dłoni widać, że nie wszyscy, którzy mienią się naukowcami,

#Papermills to psucie nauki na poziomie systemowym. Jeśli większość naukowców nie będzie dążyć do wyższych standardów etycznych, wyższej jakości i rzetelności badań, to stoimy na przegranej pozycji – mówi w wywiadzie dla FA @michaltomza.bsky.social

Sankcje USA na procesory. Eksperci oburzeni. "Polska może kupić tyle, ile pozwoli Wielki Brat" Stany Zjednoczone przedstawiły sankcje na zaawansowane procesory sztucznej inteligencji. Polska znalazła się dopiero w drugiej kategorii, obok takich krajów, jak Etiopia, Maroko, RPA, Brazylia. USA ma...

Komentarze do decyzji USA o ograniczeniu w eksporcie procesorów (w tym mój) next.gazeta.pl/next/7,15124...

„Krzysztof Walas podkreśla, że finansowy próg wejścia przy tworzeniu rozwiązań opartych na AI jest niższy niż w przypadku robotów (...). Potencjalnie otwiera to dla Polski pole do konkurowania z najlepszymi na świecie” klubjagiellonski.pl/2025/01/03/a...

😊 Happy to Share!

🎉 Our paper "Learning Graph Representation of Agent Diffusers (LGR-AD)" has been accepted as a full paper at #AAMAS (A*) International Conference on Autonomous Agents and Multiagent Systems!

#diffusion #graphs #agentsystem
@ideas-ncbr.bsky.social #WarszawUniversityOfTechnology

That was @neuripsconf.bsky.social 2024 seen through the eyes of IDEAS NCBR researchers
See more ➡️ ideas-ncbr.pl/en/ideas-ncb... #neurips

Together with my grad students and collaborators, we are launching a bi-weekly reading club about diffusion models. We'll meet every second Thursday at 15:00. Please join us in the spirit of "apes strong together" either at the Uni or online. More details in the comment

I am totally in the FOMO mode, seeing #NeurIPS24 messages. I’m not there :( this year.

However, I recommend chatting with my co-authors and students :) @bartoszpiotrowski.bsky.social @albertqjiang.bsky.social @michalnauman @mateuszostaszewski.bsky.social

2 papers on the main track and spotlight

In the gifs, see the difference between virtual robot dogs run by one of other algorithms (GIF 1) and BRO (GIF 2).

The BRO dog runs distinctly better.

💠#NeurIPS2024 is coming soon!
📜Let's focus on the BRO algorithm, introduced in 🔦spotlight paper by Michał Nauman, @mateuszostaszewski.bsky.social, Krzysztof Jankowski, @piotrmilos.bsky.social, Marek Cygan, to find out why the second dog runs better ➡️ ideas-ncbr.pl/en/bro-algor...