IDEAS NCBR
@IDEAS_NCBR
Followers
3K
Following
3K
Media
439
Statuses
2K
Scientific and research #AI centre founded by @NCBR_pl
Warsaw
Joined January 2023
Presentation of publication titled Knowledge-Driven Bayesian Uncertainty Quantification for Reliable Fake News Detection by Julia Puczyńska is being held now at the @ecai2025 (#ECAI2025) at the Nettuno Room. #ECAI2025 #AI #FakeNewsDetection
0
0
8
During @NeurIPSConf there will be an opportunity to discuss our colleague Alicja Ziarko's et. al. work titled: Contrastive Representations for Temporal Reasoning. More👉 https://t.co/VJQCZS7u7C See you at NeurIPS 2025🤝 #neurips2025 #NeurIPS #AI #IDEASNCBR
lnkd.in
This link will take you to a page that’s not on LinkedIn
10 accepted papers with our Faculty affiliation at @NeurIPSConf - the most prestigious AI&ML conference! Congrats @hbaniecki, @PrzeBiec, @marek_a_cygan, @PiotrRMilos, @mic_nau, @piotrskowron_uw, @CStanKonrad, @ewa_szczurek, @MiZawalski, @ZiarkoAlicja, S.Płotka, B.Sobieski !!!
0
0
2
Our colleague @ZiarkoAlicja had been granted @Google PhD Fellowship! 👏 Congratulations 🎉
Our doctoral student Alicja Ziarko just received the Google PhD Fellowship, presumably the most prestigious award of its kind intended for "exceptional graduate students working on innovative research in computer science and related fields" 🤩@ZiarkoAlicja
https://t.co/I0uoL4gKLk
0
0
5
Today during the @MLinPL poster session Michał Krutul presented work: Projected Compression: Trainable Projection for Efficient Transformer Compression. ( https://t.co/cBwApEpf88). The paper has been accepted to @NeurIPSConf 2025! #NeurIPS #IDEASNCBR #MLinPL
0
1
6
Today there will be a possibility to see Michał Krutul's publication "Projected Compression: Trainable Projection for Efficient Transformer Compression" during poster session. Come and join the session 🤝 #MLinPL #IDEASNCBR #AI
Last day of the Conference! After two packed days of talks and conversations, today we dive into some of the most exciting questions in machine learning. Here’s what’s happening today at the Copernicus Science Centre: 🕗 9:30 – 11:00 | Morning Sessions Main Hall • Out of
0
0
4
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
0
0
3
Tomorrow at @MLinPL our scientists will: 👥give a talk about about Limits in Causal Discovery and the Path Forward (by Mateusz Olko & Mateusz Gajewski) 🕤 9:30 📍 Hall B (CfC Session 8) 📄present a publication by Michał Krutul during poster session. 🕰️ 12:15 See you tomorrow
It’s happening: ML in PL Conference 2025 starts this Wednesday! The 9th edition of Poland’s biggest machine learning event kicks off on October 15, and we can’t wait to welcome you all to Warsaw for four days of talks, tutorials, and networking. An email has been sent to all
0
0
2
Day 2 of the ML in PL Conference 2025 is in full swing at the Copernicus Science Centre! The morning opened with sessions spanning large-scale models, robotics, and multimodal learning, from transformer compression and document AI to open-source robotics and fast inference,
0
3
5
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 👇 https://t.co/AHBHrUera0
arxiv.org
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...
0
0
1
‼️ Our colleague @ZiarkoAlicja's poblication has been accepted to #NeurIPS2025 ‼️
Can complex reasoning emerge directly from learned representations? In our new work, we study representations that capture both perceptual and temporal structure, enabling agents to reason without explicit planning. https://t.co/gGdnAUixcv
0
0
5
‼️Our colleague @apardyl's publication FlySearch: Exploring how vision-language models explore has been accepted to #NeurIPS ‼️ more about the work 👇 https://t.co/5ZzkezlwDX
arxiv.org
The real world is messy and unstructured. Uncovering critical information often requires active, goal-driven exploration. It remains to be seen whether Vision-Language Models (VLMs), which...
0
0
2
⚡️The R-GRID project, funded by @NATO_SPS, leverages #AI to strengthen power grid security. Want to learn more? Read the interview with @MacKluczynski, President of #PTBN, and Łukasz Polak from @IDEAS_NCBR. The interview was conducted by @AleksanderOlech.
Energy security and the resilience of critical infrastructure are now among the key challenges for NATO and the European Union. @AleksanderOlech
https://t.co/QD1ptxt7eY
0
4
8
W czasie konferencji odbyły się warsztaty nr. projektu R-Grid, w który zaangażowani są naukowcy IDEAS NCBR
R-GRID wspiera jubileuszową X Konferencję „Bezpieczeństwo energetyczne – filary i perspektywa rozwoju” 8–9 września 2025 r. Rzeszow University of Technology https://t.co/KjPqCnqnQX
0
2
2
A new publication by @CupiaBart ❗️👇
Almost all agentic pipelines prompt LLMs to explicitly plan before every action (ReAct), but turns out this isn't optimal for Multi-Step RL 🤔 Why? In our new work we highlight a crucial issue with ReAct and show that we should make and follow plans instead🧵
0
0
1
See you in Rzeszów where our colleague Łukasz Polak will present R-Grid - Power Grid Threat and Protection Simulator as a side event during the 10th Conference “Energy Security - Pillars and Development Perspective” #EnergySecurity #RGrid #IDEASNCBR #EnergySector #Resilience
X Konferencja "Bezpieczeństwo energetyczne - filary i perspektywa rozwoju" 📅 8–9.09.2025 r. 📍 @PolitechnikaRz 🔗 Więcej: https://t.co/VMkoF3JA4c
#IPE #InstytutPE #BezpiecznaXKonferencja #FiPR --- @motykamilosz @RuszelMariusz @dpjankowski @NATO @asseco_pl @GrupaORLEN
0
0
1
ELIAS & Bosch are inviting to a challenge. More details: https://t.co/UMLL0jdf8Y
#AI #SmartBuildings #HVAC #ELIAS #Bosch #MachineLearning #EnergyEfficency #AIChallenge #AICompetition #SustainableBuildings
0
0
0
Can complex reasoning emerge directly from learned representations? In our new work, we study representations that capture both perceptual and temporal structure, enabling agents to reason without explicit planning. https://t.co/gGdnAUixcv
4
108
774
Daman Sójka presented his work Adaptive Monocular Depth Estimation with Masked Image Consistency during Ellis Doctoral Symposium (#EDS2025). Link to the publication: https://t.co/cpOY5fC5Uu
#IDEASNCBR #AI
0
0
5