搜索结果：LDI issue brief

Timely identification of issue reports reflecting software vulnerabilities is crucial, particularly for Internet-of-Things (IoT) where analysis is slower than non-IoT systems. While Machine Learning (ML) and Large Language Models (LLMs) detect vulnerability-indicating issues in non-IoT systems, their IoT use remains unexplored. We are the first to tackle this problem by proposing two approaches: (1) combining ML and LLMs with Natural Language Processing (NLP) techniques to detect vulnerability-indicating issues of 21 Eclipse IoT projects and (2) fine-tuning a pre-trained BERT Masked Language Model (MLM) on 11,000 GitHub issues for classifying \vul. Our best performance belongs to a Support Vector Machine (SVM) trained on BERT NLP features, achieving an Area Under the receiver operator characteristic Curve (AUC) of 0.65. The fine-tuned BERT achieves 0.26 accuracy, emphasizing the importance of exposing all data during training. Our contributions set the stage for accurately detecting IoT vulnerabilities from issue reports, similar to non-IoT systems.

As retrieval-augmented generation (RAG) tackles complex tasks, increasingly expanded contexts offer richer information, but at the cost of higher latency and increased cognitive load on the model. To mitigate this bottleneck, especially for intricate multi-hop questions, we introduce BRIEF-Pro. It is a universal, lightweight compressor that distills relevant evidence for a given query from retrieved documents into a concise summary for seamless integration into in-context RAG. Using seed data consisting of relatively short contexts (fewer than 1k words), BRIEF-Pro is trained to perform abstractive compression of extended contexts exceeding 10k words across a wide range of scenarios. Furthermore, BRIEF-Pro offers flexible user control over summary length by allowing users to specify the desired number of sentences. Experiments on four open-domain multi-hop question-answering datasets show that BRIEF-Pro generates more concise and relevant summaries, enhancing performance across small, large, and proprietary language models. With the 70B reader model, 32x compression by BRIEF-Pro improves QA performance by 4.67% on average over LongLLMLingua's 9x, while requiring only 23% of its comp

We review the collaboration that led to the first rigorous proof of the stability of the hydrogen molecule within quantum mechanics and discuss several related issues concerning few-charge systems. Particular emphasis is placed on the role of symmetry breaking, the stability domains of Coulombic few-body systems, and some applications to exotic hadrons in the quark model.

Missing values are pervasive in real-world tabular data and can significantly impair downstream analysis. Imputing them is especially challenging in text-rich tables, where dependencies are implicit, complex, and dispersed across long textual fields. Recent work has explored using Large Language Models (LLMs) for data imputation, yet existing approaches typically process entire tables or loosely related contexts, which can compromise accuracy, scalability, and explainability. We introduce LDI, a novel framework that leverages LLMs through localized reasoning, selecting a compact, contextually relevant subset of attributes and tuples for each missing value. This targeted selection reduces noise, improves scalability, and provides transparent attribution by revealing the dependency relations that justify each selected attribute and the evidence behind each retrieved tuple. It makes clear not only which data influenced a prediction, but also why it was chosen. Through extensive experiments on real and synthetic datasets, we demonstrate that LDI consistently outperforms state-of-the-art imputation methods, achieving up to 8% higher accuracy with hosted LLMs and even greater gains with

Sparse approximation is important in many applications because of concise form of an approximant and good accuracy guarantees. The theory of compressed sensing, which proved to be very useful in the image processing and data sciences, is based on the concept of sparsity. A fundamental issue of sparse approximation is the problem of construction of efficient algorithms, which provide good approximation. It turns out that greedy algorithms with respect to dictionaries are very good from this point of view. They are simple in implementation and there are well developed theoretical guarantees of their efficiency. This survey/tutorial paper contains brief description of different kinds of greedy algorithms and results on their convergence and rate of convergence. Also, Chapter IV gives some typical proofs of convergence and rate of convergence results for important greedy algorithms and Chapter V gives some open problems.

The Special Issue on "Modified Gravity Approaches to the Tensions of $Λ$CDM"} in the Universe journal tackles significant challenges faced by the $Λ$CDM model, including discrepancies in the Hubble constant, growth rate of structures, and cosmological anisotropies. These issues suggest foundational cracks in the model, raising questions about the validity of General Relativity, dark energy, and cosmological principles at large scales. This collection brings together leading researchers to delve into Modified Gravity theories as potential solutions. Covering approaches from Scalar-Tensor theories to $f(R,T)$ gravity and beyond, each contribution presents innovative research aimed at addressing the limitations of the $Λ$CDM model. This Special Issue not only highlights the theoretical and empirical strengths of Modified Gravity models but also opens avenues for future investigations, emphasizing the synergy between theoretical advancements and observational evidence to deepen our cosmological understanding.

Retrieval-augmented generation (RAG) can supplement large language models (LLMs) by integrating external knowledge. However, as the number of retrieved documents increases, the input length to LLMs grows linearly, causing a dramatic increase in latency and a degradation in long-context understanding. This is particularly serious for multi-hop questions that require a chain of reasoning across documents. To accelerate inference, reduce costs, and minimize distractions, this paper presents BRIEF (Bridging Retrieval and Inference through Evidence Fusion), a lightweight approach that performs query-aware multi-hop reasoning by compressing retrieved documents into highly dense textual summaries to integrate into in-context RAG. To enable learning compression for multi-hop reasoning, we curate synthetic data by extracting atomic propositions that encapsulate distinct factoids from the source documents to compose synthetic summaries. Based on our synthetic data built entirely by open-source models, BRIEF generates more concise summaries and enables a range of LLMs to achieve exceptional open-domain question answering (QA) performance. For example, on HotpotQA, BRIEF improves the compressi

Many practitioners and academics believe in a delayed issue effect (DIE); i.e. the longer an issue lingers in the system, the more effort it requires to resolve. This belief is often used to justify major investments in new development processes that promise to retire more issues sooner. This paper tests for the delayed issue effect in 171 software projects conducted around the world in the period from 2006--2014. To the best of our knowledge, this is the largest study yet published on this effect. We found no evidence for the delayed issue effect; i.e. the effort to resolve issues in a later phase was not consistently or substantially greater than when issues were resolved soon after their introduction. This paper documents the above study and explores reasons for this mismatch between this common rule of thumb and empirical data. In summary, DIE is not some constant across all projects. Rather, DIE might be an historical relic that occurs intermittently only in certain kinds of projects. This is a significant result since it predicts that new development processes that promise to faster retire more issues will not have a guaranteed return on investment (depending on the context w

As an introduction to a Special Issue of International Journal of Theoretical and Applied Finance in Honour of the Memory of Thomas Robert Hurd we present a brief synopsis of Tom Hurd's scientific career and a bibliography of his scientific publications.

In this report, I provide a brief summary of the literature in philosophy, psychology and cognitive science about Explanatory Virtues, and link these concepts to eXplainable AI.

This special issue of the Astroparticle Physics journal is dedicated to the memory of Thomas Korff Gaisser, the Martin A. Pomerantz Professor Emeritus of Physics at the University of Delaware. Tom was one of the most prominent scientists in cosmic-ray and astroparticle physics, and also one of the founding editors of this very journal. A theoretical particle physicist by training, he dedicated his career to cosmic-ray physics. He worked on the phenomenology of high-energy particle interactions in the atmosphere, modeling extensive air showers, including the famous 'Gaisser-Hillas' function describing their longitudinal profile. A focus of his work were muon and neutrino fluxes resulting from cosmic-ray showers in the atmosphere, and their measurement with specialized experiments at the South Pole.

In this manuscript we will present a brief overview of the comorbidity concept. We will start by laying its foundations and its definitions and then describing the role that machine learning can hold in mining and defining it. The purpose of this short survey is to present a brief overview of the definition of comorbidity as a concept, and showing some of the latest applications and potentialities for the application of natural language processing and text mining techniques.

In this brief review, we report on the status of asymptotic symmetries of gravity corresponding to the class of metrices named asymptotically flat spacetimes in higher (d > 4) dimensions. We discuss the consequences of these symmetries both in classical and quantum theories. We also discuss the open issues in these aspects.

Space Very Long Baseline Interferometry is a radio astronomy technique distinguished by a record-high angular resolution reaching single-digit microseconds of arc. The paper provides a brief account of the history of developments of this technique over the period 1960s-2020s.

Exploration of the structure formation and dynamics of animate and inanimate matter on the nanometer scale is a highly interdisciplinary field of rapidly emerging research. It is relevant for various molecular and nanoscale systems of different origins and compositions and concerns numerous phenomena originating from physics, chemistry, biology, and materials science. This topical issue presents a collection of research papers devoted to different aspects of the Dynamics of Systems on the Nanoscale. Some of the contributions discuss specific applications of the research results in several modern and emerging technologies, such as controlled nanofabrication with charged particle beams or the design and practical realization of novel gamma-ray crystal-based light sources. Most works presented in this topical issue were reported at the joint Sixth International Conference "Dynamics of Systems on the Nanoscale" and the tenth International Symposium "Atomic Cluster Collisions" (DySoN-ISACC 2021), which were held in Santa Margherita Ligure, Italy, in October 2021.

Voice over IP (VoIP) is the technology allowing voice and multimedia transmissions as data packets over a private or a public IP network. Thanks to the benefits that it may provide, the VoIP technology is increasingly attracting attention and interest in the industry. Actually, VoIP allows significant benefits for customers and communication services providers such as cost savings, rich media service, phone and service portability, mobility, and the integration with other applications. Nevertheless, the deployment of the VoIP technology encounters many challenges such as architecture complexity, interoperability issues, QoS issues, and security concerns. Among these disadvantages, VoIP security issues are becoming more serious because traditional security devices, protocols, and architectures cannot adequately protect VoIP systems from recent intelligent attacks. The aim of this paper is carry out a deep analysis of the security concerns of the VoIP technology. Firstly, we present a brief overview about the VoIP technology. Then, we discuss security attacks and vulnerabilities related to VoIP protocols and devices. After that, we talk about the security profiles of the VoIP protoco

The paper includes Public Key Infrastructure (PKI), its need and requirements and introduction of some renowned PKI products. However, the major thrust of this work is that how PKI can enhance security of various systems. The paper is intended to serve as a guide on how to adequately prepare for some of the challenges that may be encountered especially in developing countries like Pakistan. The detail of PKI implementation issues is also included in the paper along with future challenges regarding implementation of PKI. Furthermore, paper includes technical issues hindering the implementation of PKI through comparison of PKI issues in Pakistan and some of Asian countries mainly Taiwan, Japan and Singapore. The paper also highlights the PKI issues and learnt lessons regarding PKI implementation and can act as a comprehensive guide for successful future PKI deployments.

Hot, massive (OB) stars experience strong line-driven stellar winds and mass loss. As the majority of efficient driving lines are metallic, the amount of wind driving and mass loss is dependent on the stellar metallicity Z. In addition, line-driven winds are intrinsically inhomogeneous and clumpy. However, to date, neither theoretical nor empirical studies of line-driven winds have investigated how such wind clumping may also depend on Z. We theoretically investigated the degree of wind clumping due to the line-deshadowing instability (LDI) as a function of Z. We performed two-dimensional hydrodynamic simulations of the LDI with an assumed one-dimensional radiation line force for a grid of O-star wind models with fixed luminosity, but with different metal contents by varying the accumulative line strength Qbar describing the total ensemble of driving lines. We find that, for this fixed luminosity, the amount of wind clumping decreases with metallicity. The decrease is clearly seen in the statistical properties of our simulations, but is nonetheless rather weak; a simple power-law fit for the dependence of the clumping factor f_cl = <rho^2>/<rho>^2 on metallicity yields

In this paper I will give a brief history of the discovery (Hamilton, 1843) of quaternions. I will address the issue of why a theory of triplets (the original goal of Hamilton) could not be developed. Finally, I will discuss briefly the history of various attempts to define holomorphic functions on quaternionic variables.