搜索结果：introduces

This paper introduces the Generative Application Firewall (GAF), a new architectural layer for securing LLM applications. Existing defenses -- prompt filters, guardrails, and data-masking -- remain fragmented; GAF unifies them into a single enforcement point, much like a WAF coordinates defenses for web traffic, while also covering autonomous agents and their tool interactions.

This 2001 paper introduces a new type of chromatic number for point sets.

This Data Descriptor introduces the dataset Enevaeldens Nyheder Online (News during Absolutism Online). The Enevaeldens Nyheder Online (ENO) dataset provides a reconstruction of the contents of major newspapers in Denmark and Norway during the period of Absolutism (1660-1849). The dataset contains approx. 474 million words, created using neural networks designed to process digitised microfilm versions of Danish newspapers as well as a smaller selection of Norwegian publications that were all hitherto illegible for computers. The contributions details this process and its results, including a way to derive standalone texts from the editions, and the accompanying BERT-model trained on a beta-version of the dataset.

We present an educational approach aimed at introducing the fundamental concepts of quantum mechanics (QM). By exploiting the formal analogy between sound wave behavior in an acoustic pipe (a drinking straw) and the quantum infinite square potential well, we provide an intuitive framework that explains the origin of energy quantization in bound quantum systems (such as atoms), introduces the concept of wave function (WF), and lays the groundwork for discussing the Heisenberg uncertainty relations. The proposed method enables low-cost experimental activities that actively involve students, facilitating a meaningful connection between theoretical principles and empirical observations. Moreover, it encourages critical reflection on the Copenhagen interpretation of the WF, promoting a deeper conceptual understanding.

This paper focuses on explaining the timbre conveyed by speech signals and introduces a task termed voice timbre attribute detection (vTAD). In this task, voice timbre is explained with a set of sensory attributes describing its human perception. A pair of speech utterances is processed, and their intensity is compared in a designated timbre descriptor. Moreover, a framework is proposed, which is built upon the speaker embeddings extracted from the speech utterances. The investigation is conducted on the VCTK-RVA dataset. Experimental examinations on the ECAPA-TDNN and FACodec speaker encoders demonstrated that: 1) the ECAPA-TDNN speaker encoder was more capable in the seen scenario, where the testing speakers were included in the training set; 2) the FACodec speaker encoder was superior in the unseen scenario, where the testing speakers were not part of the training, indicating enhanced generalization capability. The VCTK-RVA dataset and open-source code are available on the website https://github.com/vTAD2025-Challenge/vTAD.

Drawing from engineering systems and control theory, we introduce a framework to understand repository stability, which is a repository activity capacity to return to equilibrium following disturbances - such as a sudden influx of bug reports, key contributor departures, or a spike in feature requests. The framework quantifies stability through four indicators: commit patterns, issue resolution, pull request processing, and community engagement, measuring development consistency, problem-solving efficiency, integration effectiveness, and sustainable participation, respectively. These indicators are synthesized into a Composite Stability Index (CSI) that provides a normalized measure of repository health proxied by its stability. Finally, the framework introduces several important theoretical properties that validate its usefulness as a measure of repository health and stability. At a conceptual phase and open to debate, our work establishes mathematical criteria for evaluating repository stability and proposes new ways to understand sustainable development practices. The framework bridges control theory concepts with modern collaborative software development, providing a foundation

This paper introduces a novel competitive mechanism into differential evolution (DE), presenting an effective DE variant named competitive DE (CDE). CDE features a simple yet efficient mutation strategy: DE/winner-to-best/1. Essentially, the proposed DE/winner-to-best/1 strategy can be recognized as an intelligent integration of the existing mutation strategies of DE/rand-to-best/1 and DE/cur-to-best/1. The incorporation of DE/winner-to-best/1 and the competitive mechanism provide new avenues for advancing DE techniques. Moreover, in CDE, the scaling factor $F$ and mutation rate $Cr$ are determined by a random number generator following a normal distribution, as suggested by previous research. To investigate the performance of the proposed CDE, comprehensive numerical experiments are conducted on CEC2017 and engineering simulation optimization tasks, with CMA-ES, JADE, and other state-of-the-art optimizers and DE variants employed as competitor algorithms. The experimental results and statistical analyses highlight the promising potential of CDE as an alternative optimizer for addressing diverse optimization challenges.

This paper introduces a novel approach to schema inference as an on-demand function integrated directly within a DBMS, targeting NoSQL databases where schema flexibility can create challenges. Unlike previous methods relying on external frameworks like Apache Spark, our solution enables schema inference as a SQL function, allowing users to infer schemas natively within the DBMS. Implemented in Apache AsterixDB, it performs schema discovery in two phases, local inference and global schema merging, leveraging internal resources for improved performance. Experiments with real world datasets show up to a two orders of magnitude performance boost over external methods, enhancing usability and scalability.

AI workloads, particularly those driven by deep learning, are introducing novel usage patterns to high-performance computing (HPC) systems that are not comprehensively captured by standard HPC benchmarks. As one of the largest academic research centers dedicated to deep learning, Mila identified the need to develop a custom benchmarking suite to address the diverse requirements of its community, which consists of over 1,000 researchers. This report introduces Milabench, the resulting benchmarking suite. Its design was informed by an extensive literature review encompassing 867 papers, as well as surveys conducted with Mila researchers. This rigorous process led to the selection of 26 primary benchmarks tailored for procurement evaluations, alongside 16 optional benchmarks for in-depth analysis. We detail the design methodology, the structure of the benchmarking suite, and provide performance evaluations using GPUs from NVIDIA, AMD, and Intel. The Milabench suite is open source and can be accessed at github.com/mila-iqia/milabench.

This paper introduces a centralized, open-source dataset repository designed to advance NLP and NMT for Assamese, a low-resource language. The repository, available at GitHub, supports various tasks like sentiment analysis, named entity recognition, and machine translation by providing both pre-training and fine-tuning corpora. We review existing datasets, highlighting the need for standardized resources in Assamese NLP, and discuss potential applications in AI-driven research, such as LLMs, OCR, and chatbots. While promising, challenges like data scarcity and linguistic diversity remain. The repository aims to foster collaboration and innovation, promoting Assamese language research in the digital age.

In signal processing, exploring complex systems through network representations has become an area of growing interest. This study introduces the modularity graph, a new graph-based feature, to highlight the relationship across the graph communities. After showing an application to the random graph class known as Stochastic Block Model, we consider the brain functional connectivity network estimated from real EEG data. The modularity graph provides a quantitative framework for examining the interactions between neuron clusters within the brain's network. The modularity graph works alongside multiscale community detection algorithms, thereby enabling the identification of community structures at various scales. After introducing the modularity graph, we apply it to the brain functional connectivity network, estimated from publicly available EEG recordings of motor imagery experiments. Statistical analysis across multiple scales shows that the modularity graph differs for the distinct brain connectivity states associated with various motor imagery tasks. This work emphasizes the application of signal on graph processing techniques to understand brain behavior during specific cognitiv

This extended abstract introduces a class of graph learning applicable to cases where the underlying graph has polytopic uncertainty, i.e., the graph is not exactly known, but its parameters or properties vary within a known range. By incorporating this assumption that the graph lies in a polytopic set into two established graph learning frameworks, we find that our approach yields better results with less computation.

Conversational memory is the process by which humans encode, retain and retrieve verbal, non-verbal and contextual information from a conversation. Since human memory is selective, differing recollections of the same events can lead to misunderstandings and misalignments within a group. Yet, conversational facilitation systems, aimed at advancing the quality of group interactions, usually focus on tracking users' states within an individual session, ignoring what remains in each participant's memory after the interaction. Understanding conversational memory can be used as a source of information on the long-term development of social connections within a group. This paper introduces the MeMo corpus, the first conversational dataset annotated with participants' memory retention reports, aimed at facilitating computational modelling of human conversational memory. The MeMo corpus includes 31 hours of small-group discussions on Covid-19, repeated 3 times over the term of 2 weeks. It integrates validated behavioural and perceptual measures, audio, video, and multimodal annotations, offering a valuable resource for studying and modelling conversational memory and group dynamics. By intr

Nowadays, personalized recommender systems play an increasingly important role in music scenarios in our daily life with the preference prediction ability. However, existing methods mainly rely on users' implicit feedback (e.g., click, dwell time) which ignores the detailed user experience. This paper introduces Electroencephalography (EEG) signals to personal music preferences as a basis for the personalized recommender system. To realize collection in daily life, we use a dry-electrodes portable device to collect data. We perform a user study where participants listen to music and record preferences and moods. Meanwhile, EEG signals are collected with a portable device. Analysis of the collected data indicates a significant relationship between music preference, mood, and EEG signals. Furthermore, we conduct experiments to predict personalized music preference with the features of EEG signals. Experiments show significant improvement in rating prediction and preference classification with the help of EEG. Our work demonstrates the possibility of introducing EEG signals in personal music preference with portable devices. Moreover, our approach is not restricted to the music scenar

In the rapidly evolving domain of machine learning, ensuring model generalizability remains a quintessential challenge. Overfitting, where a model exhibits superior performance on training data but falters on unseen data, is a recurrent concern. This paper introduces the Overfitting Index (OI), a novel metric devised to quantitatively assess a model's tendency to overfit. Through extensive experiments on the Breast Ultrasound Images Dataset (BUS) and the MNIST dataset using architectures such as MobileNet, U-Net, ResNet, Darknet, and ViT-32, we illustrate the utility and discernment of the OI. Our results underscore the variable overfitting behaviors across architectures and highlight the mitigative impact of data augmentation, especially on smaller and more specialized datasets. The ViT-32's performance on MNIST further emphasizes the robustness of certain models and the dataset's comprehensive nature. By providing an objective lens to gauge overfitting, the OI offers a promising avenue to advance model optimization and ensure real-world efficacy.

Recent approaches have attempted to personalize dialogue systems by leveraging profile information into models. However, this knowledge is scarce and difficult to obtain, which makes the extraction/generation of profile information from dialogues a fundamental asset. To surpass this limitation, we introduce the Profile Generation Task (PGTask). We contribute with a new dataset for this problem, comprising profile sentences aligned with related utterances, extracted from a corpus of dialogues. Furthermore, using state-of-the-art methods, we provide a benchmark for profile generation on this novel dataset. Our experiments disclose the challenges of profile generation, and we hope that this introduces a new research direction.

In today's age of digital technology, ethical concerns regarding computing systems are increasing. While the focus of such concerns currently is on requirements for software, this article spotlights the hardware domain, specifically microchips. For example, the opaqueness of modern microchips raises security issues, as malicious actors can manipulate them, jeopardizing system integrity. As a consequence, governments invest substantially to facilitate a secure microchip supply chain. To combat the opaqueness of hardware, this article introduces the concept of Explainable Hardware (XHW). Inspired by and building on previous work on Explainable AI (XAI) and explainable software systems, we develop a framework for achieving XHW comprising relevant stakeholders, requirements they might have concerning hardware, and possible explainability approaches to meet these requirements. Through an exploratory survey among 18 hardware experts, we showcase applications of the framework and discover potential research gaps. Our work lays the foundation for future work and structured debates on XHW.

Set Shaping Theory, an emerging area of study, delves into the transformation of data sets via bijection functions. Central to this theory is the parameter $K$, which determines the extent of transformation, essentially reshaping the data. This article introduces the pivotal role of $K$ in the Set Shaping Theory, shedding light on its implications for data compression and transformation dynamics