At the 2025 Academia-Industry Matching Event (AIME25), held at the HUN-REN Headquarters, Dezső Miklós presented the latest results from the Rényi AI Research Group with a special focus on AI-driven healthcare innovation. The talk showcased our ongoing work in modeling the Hungarian healthcare system and developing predictive methods for understanding and forecasting patient journeys. Emphasis was placed on model interpretability and explainability—key requirements for clinical deployment—as well as our contributions to cancer screening research, including prostate and colorectal cancer.

At the open conference “The Everyday Work of State Security”, hosted at the Historical Archives of the State Security in Budapest, Rényi AI delivered a keynote presentation on its collaborative work with the Historical Archives of the Hungarian State Security and the National Archives of Hungary. The joint effort aims to accelerate archival workflows and improve the accessibility and reliability of historical collections. We introduced a trustworthy, semi-agentic Retrieval-Augmented Generation (RAG) system developed using the Council of Ministers’ meeting minutes from the first Imre Nagy government.

At the DLM Forum in Copenhagen, Rényi AI presented its joint work with the Historical Archives of the Hungarian State Security and the National Archives of Hungary to speed up archival processes and make historical collections more accessible and reliable. We showcased a trustworthy, semi-agentic Retrieval-Augmented Generation (RAG) system developed on the Council of Ministers’ meeting minutes from the first Imre Nagy government. The solution combines high-quality transcriptions with the original scans and uses a hybrid search pipeline to find the most relevant sources.

As part of the Thematic Session: AI in Medicine at the 2025 Healthy Living Symposium in Budapest (September 17–19), the Rényi AI team presents its work at the intersection of artificial intelligence and healthcare. Our talk highlights how advanced AI methods can support medical research, diagnostics, and patient care, showcasing both current projects and future directions. You can find details here.

I will introduce some theoretical results related to the convergence and generalization capabilities of neural networks, in light of articles published at the NeurIPS 2023 conference in December. The first part of the presentation will summarize some important, earlier results, and then it will cover numerous articles based on these, which were presented at NeurIPS. The aim of the presentation is to provide a comprehensive overview of the current state of the field and the currently popular research directions.

I will overview some applications of supervised and reinforcement learning methods to knot theory that might be useful in other areas of mathematics.

A reprezentációtanulás, vagyis adatokból magasabb szintű ismeret, „tudás” kinyerése a mesterséges intelligencia kutatásának egy kiemelt kérdése. Új, tudományos és társadalmi hasznosulás szempontjából is jelentős terület az önfelügyelt reprezentáció tanulás, amely a felügyelt tanuláshoz képest jóval nagyobb léptékű adatbázisokon, széles körben alkalmazható nagy modellek tanítását teszi lehetővé. Nagy nyelvi modellek tanításának mára alapeleme az önfelügyelt tanítás, és gépi látásban is több sikeres megközelítést publikáltak. Kevesebb kutatás irányul azonban arra, hogy a vizuális önfelügyelt előtanítással kialakított hálók összetett gépi látási feladatokon (pl.

Mode Combinability: Exploring Convex Combinations of Permutation Aligned Models Adrián Csiszárik, Melinda F. Kiss, Péter Kőrösi-Szabó, Márton Muntag, Gergely Papp, Dániel Varga As recently discovered (Ainsworth-Hayase-Srinivasa 2022 and others), two wide neural networks with identical network topology and trained on similar data can be permutation-aligned. That is, we can shuffle their neurons (channels) so that linearly interpolating between the two networks in parameter space becomes a meaningful operation (linear mode connectivity).

Contemporary robotics relies heavily on addressing key challenges like odometry, localization, depth perception, semantic segmentation, the creation of new viewpoints, and navigation with precision and efficiency. Implicit neural representation techniques, notably Neural Radiance Fields (NeRFs) and Generalizable NeRFs (GeNeRFs), are increasingly employed to tackle these issues. This talk focuses on exposing certain critical, but subtle flaws inherent in GeNeRFs. Adversarial attacks, while not new to various machine learning frameworks, present a significant threat.