Large Language Models (LLMs) have substantially influenced various software engineering tasks. Indeed, in the case of software refactoring, traditional LLMs have shown the ability to reduce development time and enhance code quality. However, these LLMs often rely on static, detailed instructions for specific tasks. In contrast, LLM-based agents can dynamically adapt to evolving contexts and autonomously make decisions by interacting with software tools and executing workflows. In this paper, we explore the potential of LLM-based agents in supporting refactoring activities. Specifically, we introduce RefAgent, a multi-agent LLM-based framework for end-to-end software refactoring. RefAgent consists of specialized agents responsible for planning, executing, testing, and iteratively refining refactorings using self-reflection and tool-calling capabilities. We evaluate RefAgent on eight open-source Java projects, comparing its effectiveness against a single-agent approach, a search-based refactoring tool, and historical developer refactorings. Our assessment focuses on: (1) the impact of generated refactorings on software quality, (2) the ability to identify refactoring opportunities, and (3) the contribution of each LLM agent through an ablation study. Our results show that RefAgent achieves a median unit test pass rate of 90%, reduces code smells by a median of 52.5%, and improves key quality attributes (e.g., reusability) by a median of 8.6%. Additionally, it closely aligns with developer refactorings and the search-based tool in identifying refactoring opportunities, attaining a median F1-score of 79.15% and 72.7%, respectively. Compared to single-agent approaches, RefAgent improves the median unit test pass rate by 64.7% and the median compilation success rate by 40.1%. These findings highlight the promise of multi-agent architectures in advancing automated software refactoring.

Studying the cellular architecture of the human cerebral cortex is critical for understanding brain organization and function. It requires investigating complex texture patterns in histological images, yet automatic methods that scale across whole brains are still lacking. Here we introduce CytoNet, a foundation model trained on 1 million unlabeled microscopic image patches from over 4,000 histological sections spanning ten postmortem human brains. Using co-localization in the cortical sheet for self-supervision, CytoNet encodes complex cellular patterns into expressive and anatomically meaningful feature representations. CytoNet supports multiple downstream applications, including area classification, laminar segmentation, quantification of microarchitectural variation, and data-driven mapping of previously uncharted areas. In addition, CytoNet captures microarchitectural signatures of macroscale functional organization, enabling decoding of functional network parcellations from cytoarchitectonic features. Together, these results establish CytoNet as a unified framework for scalable analysis of cortical microarchitecture and for linking cellular architecture to structure-function organization in the human cerebral cortex.

Vision-Language Models (VLMs) are increasingly deployed as autonomous agents to navigate mobile graphical user interfaces (GUIs). Operating in dynamic on-device ecosystems, which include notifications, pop-ups, and inter-app interactions, exposes them to a unique and underexplored threat vector: environmental injection. Unlike prompt-based attacks that manipulate textual instructions, environmental injection corrupts an agent's visual perception by inserting adversarial UI elements (for example, deceptive overlays or spoofed notifications) directly into the GUI. This bypasses textual safeguards and can derail execution, causing privacy leakage, financial loss, or irreversible device compromise. To systematically evaluate this threat, we introduce GhostEI-Bench, the first benchmark for assessing mobile agents under environmental injection attacks within dynamic, executable environments. Moving beyond static image-based assessments, GhostEI-Bench injects adversarial events into realistic application workflows inside fully operational Android emulators and evaluates performance across critical risk scenarios. We further propose a judge-LLM protocol that conducts fine-grained failure analysis by reviewing the agent's action trajectory alongside the corresponding screenshot sequence, pinpointing failure in perception, recognition, or reasoning. Comprehensive experiments on state-of-the-art agents reveal pronounced vulnerability to deceptive environmental cues: current models systematically fail to perceive and reason about manipulated UIs. GhostEI-Bench provides a framework for quantifying and mitigating this emerging threat, paving the way toward more robust and secure embodied agents.

We propose a method for evaluating the robustness of widely used LLM ranking systems -- variants of a Bradley--Terry model -- to dropping a worst-case very small fraction of preference data. Our approach is computationally fast and easy to adopt. When we apply our method to matchups from popular LLM ranking platforms, including Chatbot Arena and derivatives, we find that the rankings of top-performing models can be remarkably sensitive to the removal of a small fraction of preferences; for instance, dropping just 0.003% of human preferences can change the top-ranked model on Chatbot Arena. Our robustness check identifies the specific preferences most responsible for such ranking flips, allowing for inspection of these influential preferences. We observe that the rankings derived from MT-bench preferences are notably more robust than those from Chatbot Arena, likely due to MT-bench's use of expert annotators and carefully constructed prompts. Finally, we find that neither rankings based on crowdsourced human evaluations nor those based on LLM-as-a-judge preferences are systematically more sensitive than the other.

PAL Robotics, in Spain, builds a variety of Autonomous Mobile Robots (AMRs), which are deployed in diverse environments (e.g., warehouses, retail spaces, and offices), where they work alongside humans. Given that human behavior can be unpredictable and that AMRs may not have been trained to handle all possible unknown and uncertain behaviors, it is important to test AMRs under a wide range of human interactions to ensure their safe behavior. Moreover, testing in real environments with actual AMRs and humans is often costly, impractical, and potentially hazardous (e.g., it could result in human injury). To this end, we propose a Vision Language Model (VLM)-based testing approach (RVSG) for industrial AMRs developed together with PAL Robotics. Based on the functional and safety requirements, RVSG uses the VLM to generate diverse human behaviors that violate these requirements. We evaluated RVSG with several requirements and navigation routes in a simulator using the latest AMR from PAL Robotics. Our results show that, compared with the baseline, RVSG can effectively generate requirement-violating scenarios. Moreover, RVSG-generated scenarios increase variability in robot behavior, thereby helping reveal their uncertain behaviors.

Quantum Boltzmann machines (QBMs) are generative models with potential advantages in quantum machine learning, yet their training is fundamentally limited by the barren plateau problem, where gradients vanish exponentially with system size. We introduce a quantum version of the em algorithm, an information-geometric generalization of the classical Expectation-Maximization method, which circumvents gradient-based optimization on non-convex functions. Implemented on a semi-quantum restricted Boltzmann machine (sqRBM) -- a hybrid architecture with quantum effects confined to the hidden layer -- our method achieves stable learning and outperforms gradient descent on multiple benchmark datasets. These results establish a structured and scalable alternative to gradient-based training in QML, offering a pathway to mitigate barren plateaus and enhance quantum generative modeling.

Large Language Models (LLMs) have made remarkable progress, surpassing human performance on several benchmarks in domains such as mathematics and coding. A key driver of this progress has been the development of benchmark datasets. In contrast, the financial domain poses higher entry barriers due to its demand for specialized expertise, and benchmarks remain relatively scarce compared to those in mathematics or coding. We introduce EDINET-Bench, an open-source Japanese financial benchmark designed to evaluate LLMs on challenging tasks such as accounting fraud detection, earnings forecasting, and industry classification. EDINET-Bench is constructed from ten years of annual reports filed by Japanese companies. These tasks require models to process entire annual reports and integrate information across multiple tables and textual sections, demanding expert-level reasoning that is challenging even for human professionals. Our experiments show that even state-of-the-art LLMs struggle in this domain, performing only marginally better than logistic regression in binary classification tasks such as fraud detection and earnings forecasting. Our results show that simply providing reports to LLMs in a straightforward setting is not enough. This highlights the need for benchmark frameworks that better reflect the environments in which financial professionals operate, with richer scaffolding such as realistic simulations and task-specific reasoning support to enable more effective problem solving. We make our dataset and code publicly available to support future research.

This paper provides a formulation of the log-homotopy particle flow from the perspective of variational inference. We show that the transient density used to derive the particle flow follows a time-scaled trajectory of the Fisher-Rao gradient flow in the space of probability densities. The Fisher-Rao gradient flow is obtained as a continuous-time algorithm for variational inference, minimizing the Kullback-Leibler divergence between a variational density and the true posterior density. When considering a parametric family of variational densities, the function space Fisher-Rao gradient flow simplifies to the natural gradient flow of the variational density parameters. By adopting a Gaussian variational density, we derive a Gaussian approximated Fisher-Rao particle flow and show that, under linear Gaussian assumptions, it reduces to the Exact Daum and Huang particle flow. Additionally, we introduce a Gaussian mixture approximated Fisher-Rao particle flow to enhance the expressive power of our model through a multi-modal variational density. Simulations on low- and high-dimensional estimation problems illustrate our results.

Generative Large Language Models (LLMs) are increasingly used in non-generative software maintenance tasks, such as fault localization (FL). Success in FL depends on a models ability to reason about program semantics beyond surface-level syntactic and lexical features. However, widely used LLM benchmarks primarily evaluate code generation, which differs fundamentally from semantic program reasoning. Meanwhile, traditional FL benchmarks such as Defect4J and BugsInPy are either not scalable or obsolete, as their datasets have become part of LLM training data, leading to biased results. This paper presents the first large-scale empirical investigation into the robustness of LLMs fault localizability. Inspired by mutation testing, we develop an end-to-end evaluation framework that addresses key limitations in existing LLM evaluation, including data contamination, scalability, automation, and extensibility. Using real-world programs with specifications, we inject unseen faults and ask LLMs to localize them, filtering out underspecified programs where localization is ambiguous. For each successfully localized program, we apply semantic-preserving mutations (SPMs) and rerun localization to assess robustness and determine whether LLM reasoning relies on syntactic cues rather than semantics. We evaluate 10 state-of-the-art LLMs on 750,013 fault localization tasks from over 1,300 Java and Python programs. We find that SPMs cause LLMs to fail on previously localized faults in 78% of cases, and that reasoning is stronger when relevant code appears earlier in context. These results indicate that LLM code reasoning is often tied to features irrelevant to semantics. We also identify code patterns that are challenging for LLMs to reason about. Overall, our findings motivate fundamental advances in how LLMs represent, interpret, and prioritize code semantics to reason more deeply about program logic

Problem-Based Learning (PBL) has significantly impacted biomedical engineering (BME) education since its introduction in the early 2000s, effectively enhancing critical thinking and real-world knowledge application among students. With biomedical engineering rapidly converging with artificial intelligence (AI), integrating effective AI education into established curricula has become challenging yet increasingly necessary. Recent advancements, including AI's recognition by the 2024 Nobel Prize, have highlighted the importance of training students comprehensively in biomedical AI. However, effective biomedical AI education faces substantial obstacles, such as diverse student backgrounds, limited personalized mentoring, constrained computational resources, and difficulties in safely scaling hands-on practical experiments due to privacy and ethical concerns associated with biomedical data. To overcome these issues, we conducted a three-year (2021-2023) case study implementing an advanced PBL framework tailored specifically for biomedical AI education, involving 92 undergraduate and 156 graduate students from the joint Biomedical Engineering program of Georgia Institute of Technology and Emory University. Our approach emphasizes collaborative, interdisciplinary problem-solving through authentic biomedical AI challenges. The implementation led to measurable improvements in learning outcomes, evidenced by high research productivity (16 student-authored publications), consistently positive peer evaluations, and successful development of innovative computational methods addressing real biomedical challenges. Additionally, we examined the role of generative AI both as a teaching subject and an educational support tool within the PBL framework. Our study presents a practical and scalable roadmap for biomedical engineering departments aiming to integrate robust AI education into their curricula.

Robots are increasingly being deployed in public spaces such as shopping malls, sidewalks, and hospitals, where safe and socially aware navigation depends on anticipating how pedestrians respond to their presence. However, existing datasets rarely capture the full spectrum of robot-induced reactions, e.g., avoidance, neutrality, attraction, which limits progress in modeling these interactions. In this paper, we present the Pedestrian-Robot Interaction~(PeRoI) dataset that captures pedestrian motions categorized into attraction, neutrality, and repulsion across two outdoor sites under three controlled conditions: no robot present, with stationary robot, and with moving robot. This design explicitly reveals how pedestrian behavior varies across robot contexts, and we provide qualitative and quantitative comparisons to established state-of-the-art datasets. Building on these data, we propose the Neural Robot Social Force Model~(NeuRoSFM), an extension of the Social Force Model that integrates neural networks to augment inter-human dynamics with learned components and explicit robot-induced forces to better predict pedestrian motion in vicinity of robots. We evaluate NeuRoSFM by generating trajectories on multiple real-world datasets. The results demonstrate improved modeling of pedestrian-robot interactions, leading to better prediction accuracy, and highlight the value of our dataset and method for advancing socially aware navigation strategies in human-centered environments.

Graphical assets play an important role in the design and development of games. There is potential in the use of AI-driven generative tools, to aid in creating graphical assets, thus improving game design and development pipelines. However, there is little research to address how the generative methods can fit into the wider pipeline. There also no guidelines or heuristics for creating such tools. To address this gap we conducted a user study with 16 game designers and developers to examine their behaviour and interaction with generative tools for graphical assets. The findings highlight that early design stage is preferred by all participants. Designers and developers are inclined to use such tools for creating large amounts of variations at the cost of quality as they can improve the quality of the artefacts once they generate a suitable asset. The results also strongly raised the need for better integration of such tools in existing design and development environments and the need for the outputs to be in common data formats, to be manipulatable and smoothly integrate into existing environments. The study also highlights the requirement for further emphasis on the needs of the users to incorporate these tools effectively in existing pipelines. Informed by these results, we provide a set of heuristics for creating tools that meet the expectations and needs of game designers and developers.

Research in medical imaging primarily focuses on discrete data representations that poorly scale with grid resolution and fail to capture the often continuous nature of the underlying signal. Neural Fields (NFs) offer a powerful alternative by modeling data as continuous functions. While single-instance NFs have successfully been applied in medical contexts, extending them to large-scale medical datasets remains an open challenge. We therefore introduce MedFuncta, a unified framework for large-scale NF training on diverse medical signals. Building on Functa, our approach encodes data into a unified representation, namely a 1D latent vector, that modulates a shared, meta-learned NF, enabling generalization across a dataset. We revisit common design choices, introducing a non-constant frequency parameter $ω$ in widely used SIREN activations, and establish a connection between this $ω$-schedule and layer-wise learning rates, relating our findings to recent work in theoretical learning dynamics. We additionally introduce a scalable meta-learning strategy for shared network learning that employs sparse supervision during training, thereby reducing memory consumption and computational overhead while maintaining competitive performance. Finally, we evaluate MedFuncta across a diverse range of medical datasets and show how to solve relevant downstream tasks on our neural data representation. To promote further research in this direction, we release our code, model weights and the first large-scale dataset - MedNF - containing > 500 k latent vectors for multi-instance medical NFs.

Characterizing the limit behavior -- that is, the attractors -- of learning dynamics is one of the most fundamental open questions in game theory. In recent work on this front, it was conjectured that the attractors of the replicator dynamic are in one-to-one correspondence with the sink equilibria of the game -- the sink strongly connected components of a game's preference graph -- and it was established that they do stand in at least one-to-many correspondence with them. Here, we show that the one-to-one conjecture is false. We disprove this conjecture over the course of three theorems: the first disproves a stronger form of the conjecture, while the weaker form is disproved separately in the two-player and $N$-player ($N>2$) cases. By showing how the conjecture fails, we lay out the obstacles that lie ahead for characterizing attractors of the replicator, and introduce new ideas with which to tackle them. All three counterexamples derive from an object called a local source -- a point lying within the sink equilibrium, and yet which is `locally repelling'; we prove that the absence of local sources is necessary, but not sufficient, for the one-to-one property to be true. We complement this with a sufficient condition: we introduce a local property of a sink equilibrium called pseudoconvexity, and establish that when the sink equilibria of a two-player game are pseudoconvex then they precisely define the attractors. Pseudoconvexity generalizes the previous cases -- such as zero-sum games and potential games -- where this conjecture was known to hold, and reformulates these cases in terms of a simple graph property.

Many learning algorithms can be represented as Markov processes, and understanding their generalization error is a central topic in learning theory. For specific continuous-time noisy algorithms, a prominent analysis technique relies on information-theoretic tools and the so-called ``entropy flow'' method. This technique is compatible with a broad range of assumptions and leverages the convergence properties of learning dynamics to produce meaningful generalization bounds, which can also be informative or extend to discrete-time settings. Despite their success, existing entropy flow formulations are limited to specific noise and algorithm structures (\eg, Langevin dynamics). In this work, we exploit new technical tools to extend its applicability to all learning algorithms whose iterative dynamics is governed by a time-homogeneous Markov process. Our approach builds on a principled continuous-time approximation of Markov algorithms and introduces a new, exact entropy flow formula for such processes. Within this unified framework, we establish novel connections to a well-studied family of modified logarithmic Sobolev inequalities, which we use to connect the generalization error to the ergodic properties of Markov processes. Finally, we provide a detailed analysis of all the terms appearing in our theory and demonstrate its effectiveness by deriving new generalization bounds for several concrete algorithms.

Enhancing explainability in speech self-supervised learning (SSL) is important for developing reliable SSL-based speech processing systems. This study probes how speech SSL models encode speaker-specific information via a large-scale probing analysis of 11 models, decomposing identity into acoustic, prosodic, and paralinguistic attributes. The results confirm a general hierarchy wherein initial layers encode fundamental acoustics and middle layers synthesise abstract traits. Crucially, the consensus that final layers purely abstract linguistic content is challenged. It is discovered that larger models unexpectedly recover speaker identity in their deep layers. Furthermore, the intermediate representations of speech SSL models are found to capture dynamic prosody better than specialised speaker embeddings. These insights decode the complex internal mechanics of SSL models, providing guidelines for selecting interpretable and task-optimal representations.

Human drivers naturally balance the risks of different concerns while driving, including traffic rule violations, minor accidents, and fatalities. However, achieving the same behavior in autonomous driving systems remains an open problem. This paper extends a risk metric that has been verified in human-like driving studies to encompass more complex driving scenarios specified by linear temporal logic (LTL) that go beyond just collision risks. This extension incorporates the timing and severity of events into LTL specifications, thereby reflecting a human-like risk awareness. Without sacrificing expressivity for traffic rules, we adopt LTL specifications composed of safety and co-safety formulas, allowing the control synthesis problem to be reformulated as a reachability problem. By leveraging occupation measures, we further formulate a linear programming (LP) problem for this LTL-based risk metric. Consequently, the synthesized policy balances different types of driving risks, including both collision risks and traffic rule violations. The effectiveness of the proposed approach is validated by three typical traffic scenarios in Carla simulator.

Molecular simulations have assumed a paramount role in the fields of chemistry, biology, and material sciences, being able to capture the intricate dynamic properties of systems. Within this realm, coarse-grained (CG) techniques have emerged as invaluable tools to sample large-scale systems and reach extended timescales by simplifying system representation. However, CG approaches come with a trade-off: they sacrifice atomistic details that might hold significant relevance in deciphering the investigated process. Therefore, a recommended approach is to identify key CG conformations and process them using backmapping methods, which retrieve atomistic coordinates. Currently, rule-based methods yield subpar geometries and rely on energy relaxation, resulting in less-than-optimal outcomes. Conversely, machine learning techniques offer higher accuracy but are either limited in transferability between systems or tied to specific CG mappings. In this work, we introduce HEroBM, a dynamic and scalable method that employs deep equivariant graph neural networks and a hierarchical approach to achieve high-resolution backmapping. HEroBM handles any type of CG mapping, offering a versatile and efficient protocol for reconstructing atomistic structures with high accuracy. Focused on local principles, HEroBM spans the entire chemical space and is transferable to systems of varying sizes. We illustrate the versatility of our framework through diverse biological systems, including a complex real-case scenario. Here, our end-to-end backmapping approach accurately generates the atomistic coordinates of a G protein-coupled receptor bound to an organic small molecule within a cholesterol/phospholipid bilayer.

In this paper, we study zeroth-order algorithms for nonconvex minimax problems with coupled linear constraints under the deterministic and stochastic settings, which have attracted wide attention in machine learning, signal processing and many other fields in recent years, e.g., adversarial attacks in resource allocation problems and network flow problems etc. We propose two single-loop algorithms, namely the zeroth-order primal-dual alternating projected gradient (ZO-PDAPG) algorithm and the zeroth-order regularized momentum primal-dual projected gradient algorithm (ZO-RMPDPG), for solving deterministic and stochastic nonconvex-(strongly) concave minimax problems with coupled linear constraints. The iteration complexity of the two proposed algorithms to obtain an $\varepsilon$-stationary point are proved to be $\mathcal{O}(\varepsilon ^{-2})$ (resp. $\mathcal{O}(\varepsilon ^{-4})$) for solving nonconvex-strongly concave (resp. nonconvex-concave) minimax problems with coupled linear constraints under deterministic settings and $\tilde{\mathcal{O}}(\varepsilon ^{-3})$ (resp. $\tilde{\mathcal{O}}(\varepsilon ^{-6.5})$) under stochastic settings respectively. To the best of our knowledge, they are the first two zeroth-order algorithms with iterative complexity guarantees for solving nonconvex-(strongly) concave minimax problems with coupled linear constraints under the deterministic and stochastic settings. The proposed ZO-RMPDPG algorithm, when specialized to stochastic nonconvex-concave minimax problems without coupled constraints, outperforms all existing zeroth-order algorithms by achieving a better iteration complexity, thus setting a new state-of-the-art.

Mechanism design is essentially reverse engineering of games and involves inducing a game among strategic agents in a way that the induced game satisfies a set of desired properties in an equilibrium of the game. Desirable properties for a mechanism include incentive compatibility, individual rationality, welfare maximisation, revenue maximisation (or cost minimisation), fairness of allocation, etc. It is known from mechanism design theory that only certain strict subsets of these properties can be simultaneously satisfied exactly by any given mechanism. Often, the mechanisms required by real-world applications may need a subset of these properties that are theoretically impossible to be simultaneously satisfied. In such cases, a prominent recent approach is to use a deep learning based approach to learn a mechanism that approximately satisfies the required properties by minimizing a suitably defined loss function. In this paper, we present, from relevant literature, technical details of using a deep learning approach for mechanism design and provide an overview of key results in this topic. We demonstrate the power of this approach for three illustrative case studies: (a) efficient energy management in a vehicular network (b) resource allocation in a mobile network (c) designing a volume discount procurement auction for agricultural inputs. Section 6 concludes the paper.

This paper studies a linear model for multidimensional panel data of three or more dimensions with unobserved interactive fixed-effects. The main estimator uses a Neyman-orthogonal approach, and requires two preliminary steps. First, the model is embedded within a two-dimensional panel framework where factor model methods in Bai (2009) lead to consistent, but slowly converging, estimates. The second step develops a weighted-within transformation that is robust to multidimensional interactive fixed-effects and achieves the parametric rate of consistency. The estimator is shown to be asymptotically normal. The methods are implemented to estimate the demand elasticity for beer.

Why do we sometimes perceive static images as if they were moving? Visual motion illusions enjoy a sustained popularity, yet there is no definitive answer to the question of why they work. Here we present evidence in favor of the hypothesis that illusory motion is a side effect of the predictive abilities of the brain. We present a generative model, the Evolutionary Illusion GENerator (EIGen), that creates new visual motion illusions based on a video predictive neural network. We confirm that the constructed illusions are effective on human participants through a psychometric survey. Our results support the hypothesis that illusory motion might be the consequence of perceiving the brain's own predictions rather than perceiving raw visual input from the eyes. The philosophical motivation of this paper is to call attention to the untapped potential of "motivated failures", ways for artificial systems to fail as biological systems fail, as a worthy outlet for Artificial Intelligence and Artificial Life research.

The exponential growth in mobile data demand necessitates intelligent management of telecommunications infrastructure to ensure Quality of Service (QoS) and operational efficiency. This paper presents a comprehensive analysis of a multigenerational cellular network dataset, sourced from the OpenCelliD project, to identify patterns in network deployment, utilization, and infrastructure gaps. The methodology involves geographical, temporal, and performance analysis of 1,818 cell tower entries, predominantly Long Term Evolution (LTE), across three countries with a significant concentration in Pakistan. Key findings reveal the long-term persistence of legacy 2G/3G infrastructure in major urban centers, the existence of a substantial number of under-utilized towers representing opportunities for cost savings, and the identification of specific "non-4G demand zones" where active user bases are served by outdated technologies. By introducing a signal-density metric, we distinguish between absolute over-utilization and localized congestion. The results provide actionable intelligence for Mobile Network Operators (MNOs) to guide strategic LTE upgrades, optimize resource allocation, and bridge the digital divide in underserved regions.

Scaling distributed GPU training is commonly assumed to yield predictable performance gains as additional nodes are added. In practice, many large-scale deployments encounter diminishing returns and unstable behavior well before theoretical limits are reached. This paper examines why scaling fails in real systems, with a focus on the role of network and fabric effects that are often overlooked by higher-level training frameworks. We present an empirical study of distributed GPU training performance across multiple production-scale clusters. Our results show that network topology, congestion dynamics, collective synchronization behavior, and GPU locality frequently dominate end-to-end training performance once workloads move beyond a small number of nodes. Identical models and software stacks can exhibit sharply different scaling characteristics depending on fabric design and runtime communication patterns. We identify recurring failure modes that emerge as training transitions from single-node to multi-node execution, including synchronization amplification, topology-induced contention, and locality-driven performance variance. These effects are often invisible to standard profiling tools and are therefore misdiagnosed as framework or model-level inefficiencies. Based on these findings, we outline practical diagnostic principles that system builders can apply to understand scaling limits, improve predictability, and reduce the cost of large-scale distributed training.

Traditional service quality metrics often fail to capture the nuanced drivers of passenger satisfaction hidden within unstructured online feedback. This study validates a Large Language Model (LLM) framework designed to extract granular insights from such data. Analyzing over 16,000 TripAdvisor reviews for EgyptAir and Emirates (2016-2025), the study utilizes a multi-stage pipeline to categorize 36 specific service issues. The analysis uncovers a stark "operational perception disconnect" for EgyptAir: despite reported operational improvements, passenger satisfaction plummeted post-2022 (ratings < 2.0). Our approach identified specific drivers missed by conventional metrics-notably poor communication during disruptions and staff conduct-and pinpointed critical sentiment erosion in key tourism markets. These findings confirm the framework's efficacy as a powerful diagnostic tool, surpassing traditional surveys by transforming unstructured passenger voices into actionable strategic intelligence for the airline and tourism sectors.

AI agents increasingly assist with financial research, yet no benchmark evaluates their ability to retrieve specific numeric values from structured databases. We introduce FinRetrieval, a benchmark of 500 financial retrieval questions with ground truth answers, agent responses from 14 configurations across three frontier providers (Anthropic, OpenAI, Google), and complete tool call execution traces. Our evaluation reveals that tool availability dominates performance: Claude Opus achieves 90.8% accuracy with structured data APIs but only 19.8% with web search alone--a 71 percentage point gap that exceeds other providers by 3-4x. We find that reasoning mode benefits vary inversely with base capability (+9.0pp for OpenAI vs +2.8pp for Claude), explained by differences in base-mode tool utilization rather than reasoning ability. Geographic performance gaps (5.6pp US advantage) stem from fiscal year naming conventions, not model limitations. We release the dataset, evaluation code, and tool traces to enable research on financial AI systems.

The rapid growth of Retrieval-Augmented Generation (RAG) has created a proliferation of toolkits, yet a fundamental gap remains between experimental prototypes and robust, production-ready systems. We present SearchGym, a modular infrastructure designed for cross-platform benchmarking and hybrid search orchestration. Unlike existing model-centric frameworks, SearchGym decouples data representation, embedding strategies, and retrieval logic into stateful abstractions: Dataset, VectorSet, and App. This separation enables a Compositional Config Algebra, allowing designers to synthesize entire systems from hierarchical configurations while ensuring perfect reproducibility. Moreover, we analyze the "Top-$k$ Cognizance" in hybrid retrieval pipelines, demonstrating that the optimal sequence of semantic ranking and structured filtering is highly dependent on filter strength. Evaluated on the LitSearch expert-annotated benchmark, SearchGym achieves a 70% Top-100 retrieval rate. SearchGym reveals a design tension between generalizability and optimizability, presenting the potential where engineering optimization may serve as a tool for uncovering the causal mechanisms inherent in information retrieval across heterogeneous domains. An open-source implementation of SearchGym is available at: https://github.com/JeromeTH/search-gym

We find the dispersion relations of two elusive families of core-bound excitations of the Gross-Pitaevskii (GP) vortex, varicose (axisymmetric) and fluting (quadrupole) waves. For wavelengths of order the healing length, these two families -- and the well-known Kelvin wave -- possess an infinite sequence of core-bound, vortex-specific branches whose energies lie below the Bogoliubov dispersion relation. In the short-wavelength limit, these excitations can be interpreted as particles radially bound to the vortex, which acts as a waveguide. In the long-wavelength limit, the fluting waves unbind from the core, the varicose waves reduce to phonons propagating along the vortex, and the fundamental Kelvin wave is the only core-bound vortex-specific excitation. Finally, we propose a realistic spectroscopic protocol for creating and detecting the varicose wave, which we test by direct numerical simulations of the GP equation.

We introduce group surface codes, which are a natural generalization of the $\mathbb{Z}_2$ surface code, and equivalent to quantum double models of finite groups with specific boundary conditions. We show that group surface codes can be leveraged to perform non-Clifford gates in $\mathbb{Z}_2$ surface codes, thus enabling universal computation with well-established means of performing logical Clifford gates. Moreover, for suitably chosen groups, we demonstrate that arbitrary reversible classical gates can be implemented transversally in the group surface code. We present the logical operations in terms of a set of elementary logical operations, which include transversal logical gates, a means of transferring encoded information into and out of group surface codes, and preparation and readout. By composing these elementary operations, we implement a wide variety of logical gates and provide a unified perspective on recent constructions in the literature for sliding group surface codes and preparing magic states. We furthermore use tensor networks inspired by ZX-calculus to construct spacetime implementations of the elementary operations. This spacetime perspective also allows us to establish explicit correspondences with topological gauge theories. Our work extends recent efforts in performing universal quantum computation in topological orders without the braiding of anyons, and shows how certain group surface codes allow us to bypass the restrictions set by the Bravyi-K{ö}nig theorem, which limits the computational power of topological Pauli stabilizer models.

We introduce propositional team-based logics expressively complete for (quasi) downward and (quasi) upward closed properties in a syntactically dual way, by using variants of the inclusion atom. In particular, the variants of the primitive inclusion atoms used in the (quasi) upward closed setting have equivalent formulas using variants of the might modality. The duality is visible in the logics' normal forms, mirroring the duality between the (quasi) upward and downward closed settings, where the quasi variants take special care of the empty and full team. Furthermore, we defined sound and complete natural deduction systems for each logic.