AIhub monthly digest: August 2026 - IJCAI-ECAI in Bremen, the mathematics of simplicity, and does AI change the way we think? Welcome to our monthly digest, where you can catch up with any AIhub stories you may have missed, peruse the latest news, recap recent events, and more. This month, we report on events at IJCAI-ECAI 2026, learn about the mathematics of simplicity, investigate the accountability vacuum, and find out how AI changes the way we think. On Rashomon sets, the mathematics of simplicity, and why we don't need black boxes: an interview with Cynthia Rudin In the latest in our series of interviews with AI pioneers, we hear from Cynthia Rudin about interpretability, noise, and the case against complexity for complexity's sake. The 35th International Joint Conference on Artificial Intelligence and the 29th European Conference on Artificial Intelligence (IJACI-ECAI 2026) was held from 15-21 August, in Bremen, Germany.

RoboCup 2026 saw history made, as two teams of 11 humanoids took to the soccer field, the first time a full complement of robots has competed. The game saw (Bremen, Germany) take on (Leipzig, Germany), with both sides using machines designed by Booster Robotics. Back in 1997, RoboCup's founders set the lofty goal of developing a team of autonomous robots that could beat the human World Cup champions by 2050. There has been significant progress since those early days, and this match saw another step towards that ambition. "This match shows how far humanoid robotics has come," said Ubbo Visser, President of the RoboCup Federation.

The 35th International Joint Conference on Artificial Intelligence and the 29th European Conference on Artificial Intelligence (IJACI-ECAI 2026) took place in Bremen, Germany from 15-21 August. In the second of our social media round-ups we find out about the doctoral consortium, drop into the keynotes, and look back on an action-packed week. If you missed it, you can catch our round-up part 1 here . Today: presented my work, LLM Parametric Knowledge Gaps as a Democratic Harm, at the Augmented Democracies Workshop, IJCAI-ECAI 2026. Jaap Jumelet, UvA, was awarded the 2025 EurAI Dissertation Award for his outstanding PhD thesis Finding Structure in Language Models by the EurAI Board member and 2025 EurAI Disseration Award Chair Fredrik Heintz, https://t.co/Uo8YHTNSRK

The 35th International Joint Conference on Artificial Intelligence and the 29th European Conference on Artificial Intelligence (IJACI-ECAI 2026) is currently taking place in Bremen, Germany. The first couple of days saw the attendees enjoy some of the many tutorials and workshops on offer. The official opening ceremony took place yesterday morning (Monday 17 August). Find out what the participants have been getting up to so far. Toby Walsh, UNSW, IJCAI-ECAI 2026 Ethics Co-Chair, will deliver his AI Lecture, on Mon, Aug 17. Open to the general public, it offers a preview of themes explored in his forthcoming book, God AI: Boom or Doom?

The 35th International Joint Conference on Artificial Intelligence and the 29th European Conference on Artificial Intelligence (IJACI-ECAI 2026) will be held from 15-21 August, in Bremen, Germany. The conference will feature workshops and tutorials, keynote and invited talks, technical presentations, posters, diversity and inclusion event, outreach, and more. Experts from research, practice, public institutions and civil society discuss how AI is changing key areas of everyday life in moderated panel discussions. The AI Lounges will be held in German. The tutorials will also take place from 15-17 August.

Many normalizing flow architectures impose regularity constraints, yet their distributional approximation properties are not fully characterized. We study the expressivity of bi-Lipschitz normalizing flows through the lens of score-based diffusion models. For the probability flow ODE of a variance-preserving diffusion, Lipschitz regularity of the score induces a flow of bi-Lipschitz diffeomorphic transport maps. This ODE bridge allows us to analyze the distributional approximation power of bi-Lipschitz normalizing flows and, conversely, derive deterministic convergence guarantees for diffusion-based transport. Our key idea is to use the probability flow ODE to link regularity of the score to regularity of the induced transport maps. We verify score regularity for broad target densities, including compactly supported densities, Gaussian convolutions of compactly supported measures and finite Gaussian mixtures. We obtain a universal distributional approximation result: Gaussian pullbacks induced by bi-Lipschitz variance-preserving transport maps are $L^1$-dense among all probability densities. For Gaussian convolution targets, we further obtain convergence in Kullback-Leibler divergence without early stopping.

Deep learning (DL) has become a cornerstone of modern machine learning (ML) praxis. We introduce the R package mlr3torch, which is an extensible DL framework for the mlr3 ecosystem. It is built upon the torch package, and simplifies the definition, training, and evaluation of neural networks for both tabular data and generic tensors (e.g., images) for classification and regression. The package implements predefined architectures, and torch models can easily be converted to mlr3 learners. It also allows users to define neural networks as graphs. This representation is based on the graph language defined in mlr3pipelines and allows users to define the entire modeling workflow, including preprocessing, data augmentation, and network architecture, in a single graph. Through its integration into the mlr3 ecosystem, the package allows for convenient resampling, benchmarking, preprocessing, and more. We explain the package's design and features and show how to customize and extend it to new problems. Furthermore, we demonstrate the package's capabilities using three use cases, namely hyperparameter tuning, fine-tuning, and defining architectures for multimodal data. Finally, we present some runtime benchmarks.

We consider the problem of conditional independence (CI) testing and adopt a kernel-based approach. Kernel-based CI tests embed variables in reproducing kernel Hilbert spaces, regress their embeddings on the conditioning variables, and test the resulting residuals for marginal independence. This approach yields tests that are sensitive to a broad range of conditional dependencies. Existing methods, however, rely heavily on kernel ridge regression, which is computationally expensive when properly tuned and yields poorly calibrated tests when left untuned, which limits their practical usefulness. We propose the Generalised Kernel Covariance Measure (GKCM), a regression-model-agnostic kernel-based CI test that accommodates a broad class of regression estimators. Building on the Generalised Hilbertian Covariance Measure framework (Lundborg et al., 2022), we characterise conditions under which GKCM satisfies uniform asymptotic level guarantees. In simulations, GKCM paired with tree-based regression models frequently outperforms state-of-the-art CI tests across a diverse range of data-generating processes, achieving better type I error control and competitive or superior power.

Thus, weaker models that give imprecise results quickly can be advantageous, provided inaccuracies can be resolved using few queries to a stronger model. In the fundamental problem of computing a maximum-weight basis of a matroid, a well-known generalization of many combinatorial optimization problems, algorithms have access to a clean oracle to query matroid information. We additionally equip algorithms with a fast but dirty oracle. We design and analyze practical algorithms that only use few clean queries w.r.t. the quality of the dirty oracle, while maintaining robustness against arbitrarily poor dirty oracles, approaching the performance of classic algorithms for the given problem. Notably, we prove that our algorithms are, in many respects, best-possible. Further, we outline extensions to other matroid oracle types, non-free dirty oracles and other matroid problems.