Large language models (LLMs) can generate programs that pass unit tests, but passing tests does not guarantee reliable runtime behavior. We find that different correct solutions to the same task can show very different memory and performance patterns, which can lead to hidden operational risks. We present a framework to measure execution-time memory stability across multiple correct generations. At the solution level, we introduce Dynamic Mean Pairwise Distance (DMPD), which uses Dynamic Time Warping to compare the shapes of memory-usage traces after converting them into Monotonic Peak Profiles (MPPs) to reduce transient noise. Aggregating DMPD across tasks yields a model-level Model Instability Score (MIS). Experiments on BigOBench and CodeContests show substantial runtime divergence among correct solutions. Instability often increases with higher sampling temperature even when pass@1 improves. We also observe correlations between our stability measures and software engineering indicators such as cognitive and cyclomatic complexity, suggesting links between operational behavior and maintainability. Our results support stability-aware selection among passing candidates in CI/CD to
Existing on-device AI architectures for resource-constrained environments face two critical limitations: they lack compactness, with parameter requirements scaling proportionally to task complexity, and they exhibit poor generalizability, performing effectively only on specific application domains (e.g., models designed for regression tasks cannot adapt to natural language processing (NLP) applications). In this paper, we propose CURA, an architecture inspired by analog audio signal processing circuits that provides a compact and lightweight solution for diverse machine learning tasks across multiple domains. Our architecture offers three key advantages over existing approaches: (1) Compactness: it requires significantly fewer parameters regardless of task complexity; (2) Generalizability: it adapts seamlessly across regression, classification, complex NLP, and computer vision tasks; and (3) Complex pattern recognition: it can capture intricate data patterns while maintaining extremely low model complexity. We evaluated CURA across diverse datasets and domains. For compactness, it achieved equivalent accuracy using up to 2,500 times fewer parameters compared to baseline models. For
This paper reviews the current state and emerging trends in synthetic speech detection. It outlines the main data-driven approaches, discusses the advantages and drawbacks of focusing future research solely on neural encoding detection, and offers recommendations for promising research directions. Unlike works that introduce new detection methods or datasets, this paper aims to guide future state-of-the-art research in the field and to highlight the risk of overcommitting to approaches that may not stand the test of time.
Diffusion model have been successfully applied to many inverse problems, including MRI and CT reconstruction. Researchers typically re-purpose models originally designed for unconditional sampling without modifications. Using two different posterior sampling algorithms, we show empirically that such large networks are not necessary. Our smallest model, effectively a ResNet, performs almost as good as an attention U-Net on in-distribution reconstruction, while being significantly more robust towards distribution shifts. Furthermore, we introduce models trained on natural images and demonstrate that they can be used in both MRI and CT reconstruction, out-performing model trained on medical images in out-of-distribution cases. As a result of our findings, we strongly caution against simply re-using very large networks and encourage researchers to adapt the model complexity to the respective task. Moreover, we argue that a key step towards a general diffusion-based prior is training on natural images.
In this paper, we study some codes of algebraic geometry related to certain maximal curves. Quantum stabilizer codes obtained through the self orthogonality of Hermitian codes of this error correcting do not always have good parameters. However, appropriate parameters found that the Hermitian self-orthogonal code quantum stabilizer code has good parameters. Therefore, we investigated the quantum stabilizer code at a certain maximum curve and modified its parameters. Algebraic geometry codes show promise for enabling high data rate transmission over noisy power line communication channels.
Many datasets represent a combination of different ways of looking at the same data that lead to different generalizations. For example, a corpus with examples generated by different people may be mixtures of many perspectives and can be viewed with different perspectives by others. It isnt always possible to represent the viewpoints by a clean separation, in advance, of examples representing each viewpoint and train a separate model for each viewpoint. We introduce lensing, a mixed initiative technique to extract lenses or mappings between machine learned representations and perspectives of human experts, and to generate lensed models that afford multiple perspectives of the same dataset. We apply lensing for two classes of latent variable models: a mixed membership model, a matrix factorization model in the context of two mental health applications, and we capture and imbue the perspectives of clinical psychologists into these models. Our work shows the benefits of the machine learning practitioner formally incorporating the perspective of a knowledgeable domain expert into their models rather than estimating unlensed models themselves in isolation.
A mysterious cosmic explosion has astronomers buzzing, as a strange event may hint at an entirely new kind of stellar cataclysm。 After detecting ripples in space-time, scientists spotted a fast-fading red glow that initially looked like a rare kilonova—the kind of collision that forges gold and uranium。 But just days later, the signal shifted, beha
A new AI-driven method called GOFLOW is turning weather satellite images into highly detailed maps of ocean currents。 By tracking how temperature patterns shift over time, it can reveal fast-moving, small-scale currents that were previously impossible to observe directly。 These currents are key to understanding climate, marine ecosystems, and carbo
After two centuries of failed attempts, scientists have finally grown dolomite in the lab, cracking a long-standing geological puzzle。 They discovered that the mineral’s growth stalls because of tiny defects—but in nature, those flaws get washed away over time。 By mimicking this process with precise simulations and electron beam pulses, the team ac
Scientists have created a powerful new way to control quantum systems, achieving the first-ever demonstration of quadsqueezing—an elusive fourth-order quantum effect。 By combining simple forces in a clever way, they made previously hidden quantum behaviors visible and usable, opening new frontiers for quantum technology
A group of undergraduate students pulled off something remarkable: they built their own dark matter detector and used it to probe one of physics’ biggest mysteries。 Working with limited resources but plenty of creativity, they designed a stripped-down experiment to hunt for axions — hypothetical particles that could make up dark matter
A massive cosmic milestone has just been reached: scientists have completed the largest high-resolution 3D map of the universe ever created。 Built using data from over 47 million galaxies and quasars, this map could unlock new clues about dark energy—the mysterious force driving the universe’s expansion。 Despite setbacks like wildfire disruptions,
Quantum physics once shocked scientists by revealing that particles can behave like waves—and now, that strange behavior has been pushed even further。 For the first time, researchers have observed wave-like interference in positronium, an exotic “atom” made of an electron and its antimatter partner, a positron。 This breakthrough not only strengthen
Engineers at Northwestern University have taken a striking leap toward merging machines with the human brain by printing artificial neurons that can actually communicate with real ones。 These flexible, low-cost devices generate lifelike electrical signals capable of activating living brain cells, a breakthrough demonstrated in mouse brain tissue