Audio comprehension—including speech, non-speech sounds, and music—is essential for achieving human-level intelligence. Consequently, AI agents must demonstrate holistic audio understanding to qualify as generally intelligent. However, evaluating auditory intelligence comprehensively remains challenging. To address this gap, we introduce MMAU-Pro, the most comprehensive and rigorously curated benchmark for assessing audio intelligence in AI systems. MMAU-Pro contains 5,305 instances, where each instance has one or more audios paired with human expert-generated question-answer pairs, spanning speech, sound, music, and their combinations. Unlike existing benchmarks, MMAU-Pro evaluates auditory intelligence across 49 unique skills and multiple complex dimensions, including long-form audio comprehension, spatial audio reasoning, multi-audio understanding, among others. All questions are meticulously designed to require deliberate multi-hop reasoning, including both multiple-choice and open-ended response formats. Importantly, audio data is sourced directly ``from the wild" rather than from existing datasets with known distributions. We evaluate 22 leading open-source and proprietary multimodal AI models, revealing significant limitations: even state-of-the-art models such as Gemini 2.5 Flash and Audio Flamingo 3 achieve only 57.33% and 45.9% accuracy, respectively, approaching random performance in multiple categories. Our extensive analysis highlights specific shortcomings and provides novel insights, offering actionable perspectives for the community to enhance future AI systems' progression toward audio general intelligence.
Submodular maximization is a classic algorithmic problem with multiple applications in data mining and machine learning; there, the growing need to deal with massive instances motivates the design of algorithms balancing the quality of the solution with applicability. For the latter, an important measure is the adaptive complexity, which captures the number of sequential rounds of parallel computation needed by an algorithm to terminate. In this work we obtain the first constant factor approximation algorithm for non-monotone submodular maximization subject to a knapsack constraint with near-optimal $O(\log n)$ adaptive complexity. Low adaptivity by itself, however, is not enough: a crucial feature to account for is represented by the total number of function evaluations (or value queries). Our algorithm asks $\tilde{O}(n^2)$ value queries, but can be modified to run with only $\tilde{O}(n)$ instead, while retaining a low adaptive complexity of $O(\log^2n)$. Besides the above improvement in adaptivity, this is also the first combinatorial approach with sublinear adaptive complexity for the problem and yields algorithms comparable to the state-of-the-art even for the special cases of cardinality constraints or monotone objectives.
Programmer attribution seeks to identify or verify the author of a source code artifact using stylistic, structural, or behavioural characteristics. This problem has been studied across software engineering, security, and digital forensics, resulting in a growing and methodologically diverse set of publications. This paper presents a systematic mapping study of programmer attribution research focused on source code analysis. From an initial set of 135 candidate publications, 47 studies published between 2012 and 2025 were selected through a structured screening process. The included works are analysed along several dimensions, including authorship tasks, feature categories, learning and modelling approaches, dataset sources, and evaluation practices. Based on this analysis, we derive a taxonomy that relates stylistic and behavioural feature types to commonly used machine learning techniques and provide a descriptive overview of publication trends, benchmarks, programming languages. A content-level analysis highlights the main thematic clusters in the field. The results indicate a strong focus on closed-world authorship attribution using stylometric features and a heavy reliance on a small number of benchmark datasets, while behavioural signals, authorship verification, and reproducibility remain less explored. The study consolidates existing research into a unified framework and outlines methodological gaps that can guide future work. This manuscript is currently under review. The present version is a preprint.
The widespread deployment of machine learning systems in critical real-world decision-making applications has highlighted the urgent need for counterfactual explainability methods that operate effectively. Global counterfactual explanations, expressed as actions to offer recourse, aim to provide succinct explanations and insights applicable to large population subgroups. High effectiveness, measured by the fraction of the population that is provided recourse, ensures that the actions benefit as many individuals as possible. Keeping the cost of actions low ensures the proposed recourse actions remain practical and actionable. Limiting the number of actions that provide global counterfactuals is essential to maximize interpretability. The primary challenge, therefore, is to balance these trade-offs—maximizing effectiveness, minimizing cost, while maintaining a small number of actions. We introduce GLANCE, a versatile and adaptive algorithm that employs a novel agglomerative approach, jointly considering both the feature space and the space of counterfactual actions, thereby accounting for the distribution of points in a way that aligns with the model's structure. This design enables the careful balancing of the trade-offs among the three key objectives, with the size objective functioning as a tunable parameter to keep the actions few and easy to interpret. Our extensive experimental evaluation demonstrates that GLANCE consistently shows greater robustness and performance compared to existing methods across various datasets and models.
Let {Λ_n={λ_1,n,…,λ_d_n,n}}_n be a sequence of finite multisets of real numbers such that d_n→∞ as n→∞, and let f:Ω⊂ℝ^d→ℝ be a Lebesgue measurable function defined on a domain Ω with 0<μ_d(Ω)<∞, where μ_d is the Lebesgue measure in ℝ^d. We say that {Λ_n}_n has an asymptotic distribution described by f, and we write {Λ_n}_n∼ f, if lim_n→∞1/d_n∑_i=1^d_nF(λ_i,n)=1/μ_d(Ω)∫_Ω F(f( x)) d x (*) for every continuous function F with bounded support. If Λ_n is the spectrum of a matrix A_n, we say that {A_n}_n has an asymptotic spectral distribution described by f and we write {A_n}_n∼_λ f. In the case where d=1, Ω is a bounded interval, Λ_n⊆ f(Ω) for all n, and f satisfies suitable conditions, Bogoya, Böttcher, Grudsky, and Maximenko proved that the asymptotic distribution (*) implies the uniform convergence to 0 of the difference between the properly sorted vector [λ_1,n,…,λ_d_n,n] and the vector of samples [f(x_1,n),…,f(x_d_n,n)], i.e., lim_n→∞ max_i=1,…,d_n|f(x_i,n)-λ_τ_n(i),n|=0, (**) where x_1,n,…,x_d_n,n is a uniform grid in Ω and τ_n is the sorting permutation. We extend this result to the case where d≥1 and Ω is a Peano–Jordan measurable set (i.e., a bounded set with μ_d(∂Ω)=0). See the rest of the abstract in the manuscript.
