Of the 1,104 papers initially retrieved, only ten met the IC. This process was conducted by the main author and subsequently reviewed by the other authors. Secondly, we applied backward and forward snowballing (via Google Scholar) on each selected paper to identify additional relevant studies not captured by the initial Scopus search. This iterative process is illustrated in Figure 2. Through snowballing, we identified 17 additional papers, bringing the total number of selected studies to 27. In total, we screened over 2,400 papers across the Scopus search and snowballing iterations. D. Data Extraction and Classification Scheme The Data Extraction and Classification Scheme for each paper is outlined in Table II. The selection process and the extracted data are documented in our online Zenodo repository, which includes information on each identified paper, the reason for its inclusion or exclusion, and the data extracted to answer each RQ.

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 hear about explainable AI for robotics, explore privacy-preserving generative models, and find out what RoboCup 2025 has in store. RoboCup is an international scientific initiative with the goal of advancing the state of the art of intelligent robots, AI and automation. The annual RoboCup event, where teams gather from across the globe to take part in competitions across a number of leagues, will this year take place in Brazil, from 15-21 July. We spoke to Marco Simões, one of the General Chairs of RoboCup 2025 and President of RoboCup Brazil, to find out what plans they have for the event, some new initiatives, and how RoboCup has grown in Brazil over the past ten years.

Large Language Models (LLMs) ) have demonstrated promise in boosting productivity across AI-powered tools, yet existing benchmarks like Massive Multitask Language Understanding (MMLU) inadequately assess enterprise-specific task complexities. We propose a 14-task framework grounded in Bloom's Taxonomy to holistically evaluate LLM capabilities in enterprise contexts. To address challenges of noisy data and costly annotation, we develop a scalable pipeline combining LLM-as-a-Labeler, LLM-as-a-Judge, and corrective retrieval-augmented generation (CRAG), curating a robust 9,700-sample benchmark. Evaluation of six leading models shows open-source contenders like DeepSeek R1 rival proprietary models in reasoning tasks but lag in judgment-based scenarios, likely due to overthinking. Our benchmark reveals critical enterprise performance gaps and offers actionable insights for model optimization. This work provides enterprises a blueprint for tailored evaluations and advances practical LLM deployment.

We introduce CogGen, a learner-centered AI architecture that transforms programming videos into interactive, adaptive learning experiences by integrating student modeling with generative AI tutoring based on the Cognitive Apprenticeship framework. The architecture consists of three components: (1) video segmentation by learning goals, (2) a conversational tutoring engine applying Cognitive Apprenticeship strategies, and (3) a student model using Bayesian Knowledge Tracing to adapt instruction. Our technical evaluation demonstrates effective video segmentation accuracy and strong pedagogical alignment across knowledge, method, action, and interaction layers. Ablation studies confirm the necessity of each component in generating effective guidance. This work advances AI-powered tutoring by bridging structured student modeling with interactive AI conversations, offering a scalable approach to enhancing video-based programming education.

Large language models (LLMs) predominantly use autoregressive (AR) approaches, but masked diffusion models (MDMs) are emerging as viable alternatives. A key challenge in comparing AR and MDM paradigms is their typical architectural difference: AR models are often decoder-only, while MDMs have largely been encoder-only. This practice of changing both the modeling paradigm and architecture simultaneously makes direct comparisons unfair, as it's hard to distinguish whether observed differences stem from the paradigm itself or the architectural shift. This research evaluates MDMs within a decoder-only framework to: (1) equitably compare MDM (as Any-Order AR, or AO-AR) and standard AR paradigms. Our investigation suggests that the standard AO-AR objective, which averages over all token permutations, may benefit from refinement, as many permutations appear less informative compared to the language's inherent left-to-right structure. (2) Investigate architectural influences (decoder-only vs. encoder-only) within MDMs. We demonstrate that while encoder-only MDMs model a simpler conditional probability space, decoder-only MDMs can achieve dramatic generation speedups ($\sim25\times$) and comparable perplexity with temperature annealing despite modeling a vastly larger space, highlighting key trade-offs. This work thus decouples core paradigm differences from architectural influences, offering insights for future model design. Code is available at https://github.com/scxue/AO-GPT-MDM.

The interface between stochastic analysis and machine learning is a rapidly evolving field, with path signatures - iterated integrals that provide faithful, hierarchical representations of paths - offering a principled and universal feature map for sequential and structured data. Rooted in rough path theory, path signatures are invariant to reparameterization and well-suited for modelling evolving dynamics, long-range dependencies, and irregular sampling - common challenges in real-world time series and graph data. This thesis investigates how to harness the expressive power of path signatures within scalable machine learning pipelines. It introduces a suite of models that combine theoretical robustness with computational efficiency, bridging rough path theory with probabilistic modelling, deep learning, and kernel methods. Key contributions include: Gaussian processes with signature kernel-based covariance functions for uncertainty-aware time series modelling; the Seq2Tens framework, which employs low-rank tensor structure in the weight space for scalable deep modelling of long-range dependencies; and graph-based models where expected signatures over graphs induce hypo-elliptic diffusion processes, offering expressive yet tractable alternatives to standard graph neural networks. Further developments include Random Fourier Signature Features, a scalable kernel approximation with theoretical guarantees, and Recurrent Sparse Spectrum Signature Gaussian Processes, which combine Gaussian processes, signature kernels, and random features with a principled forgetting mechanism for multi-horizon time series forecasting with adaptive context length. We hope this thesis serves as both a methodological toolkit and a conceptual bridge, and provides a useful reference for the current state of the art in scalable, signature-based learning for sequential and structured data.

[Context] The increasing adoption of machine learning (ML) in software systems demands specialized ideation approaches that address ML-specific challenges, including data dependencies, technical feasibility, and alignment between business objectives and probabilistic system behavior. Traditional ideation methods like Lean Inception lack structured support for these ML considerations, which can result in misaligned product visions and unrealistic expectations. [Goal] This paper presents Define-ML, a framework that extends Lean Inception with tailored activities - Data Source Mapping, Feature-to-Data Source Mapping, and ML Mapping - to systematically integrate data and technical constraints into early-stage ML product ideation. [Method] We developed and validated Define-ML following the Technology Transfer Model, conducting both static validation (with a toy problem) and dynamic validation (in a real-world industrial case study). The analysis combined quantitative surveys with qualitative feedback, assessing utility, ease of use, and intent of adoption. [Results] Participants found Define-ML effective for clarifying data concerns, aligning ML capabilities with business goals, and fostering cross-functional collaboration. The approach's structured activities reduced ideation ambiguity, though some noted a learning curve for ML-specific components, which can be mitigated by expert facilitation. All participants expressed the intention to adopt Define-ML. [Conclusion] Define-ML provides an openly available, validated approach for ML product ideation, building on Lean Inception's agility while aligning features with available data and increasing awareness of technical feasibility.

"When I think about the future of robots and society, I don't see machine overlords" Are you worried that AI-powered robots are going to steal our jobs and maybe kill us all? But it is time to play devil's advocate with yourself and consider whether the opposite might be true. My new novel, Automatic Noodle, out later this year, is about four robots who struggle to find employment in a country where humans have made laws preventing bots from unionising, opening bank accounts, voting and owning their own businesses. Yes, it is science fiction. But it is based on real tech – and, more importantly, it explores the implications of our deeply held suspicion that robots are evil.

RoboCup is an international scientific initiative with the goal of advancing the state of the science of intelligent robots, AI and automation. The annual RoboCup event will take place from 15-21 July in Salvador, Brazil. The RoboCupRescue League is an important element of the competition and focuses on the challenges involved in search and rescue applications. We caught up with Adam Jacoff, co-founder of the RoboCupRescue league, former RoboCup Trustee, and chair of the organising committee, to find out more. The RoboCupRescue League is now in its 25th year hosting competitions and workshops all around the world.

Automatic Speech Recognition (ASR) has reached impressive accuracy for high-resource languages, yet its utility in linguistic fieldwork remains limited. Recordings collected in fieldwork contexts present unique challenges, including spontaneous speech, environmental noise, and severely constrained datasets from under-documented languages. In this paper, we benchmark the performance of two fine-tuned multilingual ASR models, MMS and XLS-R, on five typologically diverse low-resource languages with control of training data duration. Our findings show that MMS is best suited when extremely small amounts of training data are available, whereas XLS-R shows parity performance once training data exceed one hour. We provide linguistically grounded analysis for further provide insights towards practical guidelines for field linguists, highlighting reproducible ASR adaptation approaches to mitigate the transcription bottleneck in language documentation.