Search

This paper explores the advancements in making large language models (LLMs) more human-like. We focus on techniques that enhance natural language understanding, conversational coherence, and emotional intelligence in AI systems. The study evaluates various approaches, including fine-tuning with diverse datasets, incorporating psychological principles, and designing models that better mimic human reasoning patterns. Our findings demonstrate that these enhancements not only improve user interactions but also open new possibilities for AI applications across different domains. Future work will address the ethical implications and potential biases introduced by these human-like attributes.

The rapid proliferation of Industrial Internet of Things (IIoT) systems necessitates advanced, interpretable, and scalable intrusion detection systems (IDS) to combat emerging cyber threats. Traditional IDS face challenges such as high computational demands, limited explainability, and inflexibility against evolving attack patterns. To address these limitations, this study introduces the Lightweight Explainable Network Security framework (LENS-XAI), which combines robust intrusion detection with enhanced interpretability and scalability. LENS-XAI integrates knowledge distillation, variational autoencoder models, and attribution-based explainability techniques to achieve high detection accuracy and transparency in decision-making. By leveraging a training set comprising 10% of the available data, the framework optimizes computational efficiency without sacrificing performance. Experimental evaluation on four benchmark datasets: Edge-IIoTset, UKM-IDS20, CTU-13, and NSL-KDD, demonstrates the framework's superior performance, achieving detection accuracies of 95.34%, 99.92%, 98.42%, and 99.34%, respectively. Additionally, the framework excels in reducing false positives and adapting to complex attack scenarios, outperforming existing state-of-the-art methods. Key strengths of LENS-XAI include its lightweight design, suitable for resource-constrained environments, and its scalability across diverse IIoT and cybersecurity contexts. Moreover, the explainability module enhances trust and transparency, critical for practical deployment in dynamic and sensitive applications. This research contributes significantly to advancing IDS by addressing computational efficiency, feature interpretability, and real-world applicability. Future work could focus on extending the framework to ensemble AI systems for distributed environments, further enhancing its robustness and adaptability.

In the evolving landscape of autonomous vehicles, ensuring robust in-vehicle network (IVN) security is paramount. This paper introduces an advanced intrusion detection system (IDS) called KD-XVAE that uses a Variational Autoencoder (VAE)-based knowledge distillation approach to enhance both performance and efficiency. Our model significantly reduces complexity, operating with just 1669 parameters and achieving an inference time of 0.3 ms per batch, making it highly suitable for resource-constrained automotive environments. Evaluations in the HCRL Car-Hacking dataset demonstrate exceptional capabilities, attaining perfect scores (Recall, Precision, F1 Score of 100%, and FNR of 0%) under multiple attack types, including DoS, Fuzzing, Gear Spoofing, and RPM Spoofing. Comparative analysis on the CICIoV2024 dataset further underscores its superiority over traditional machine learning models, achieving perfect detection metrics. We furthermore integrate Explainable AI (XAI) techniques to ensure transparency in the model's decisions. The VAE compresses the original feature space into a latent space, on which the distilled model is trained. SHAP(SHapley Additive exPlanations) values provide insights into the importance of each latent dimension, mapped back to original features for intuitive understanding. Our paper advances the field by integrating state-of-the-art techniques, addressing critical challenges in the deployment of efficient, trustworthy, and reliable IDSes for autonomous vehicles, ensuring enhanced protection against emerging cyber threats.

This paper proposes Pix2Next, a novel image-to-image translation framework designed to address the challenge of generating high-quality Near-Infrared (NIR) images from RGB inputs. Our approach leverages a state-of-the-art Vision Foundation Model (VFM) within an encoder-decoder architecture, incorporating cross-attention mechanisms to enhance feature integration. This design captures detailed global representations and preserves essential spectral characteristics, treating RGB-to-NIR translation as more than a simple domain transfer problem. A multi-scale PatchGAN discriminator ensures realistic image generation at various detail levels, while carefully designed loss functions couple global context understanding with local feature preservation. We performed experiments on the RANUS dataset to demonstrate Pix2Next's advantages in quantitative metrics and visual quality, improving the FID score by 34.81% compared to existing methods. Furthermore, we demonstrate the practical utility of Pix2Next by showing improved performance on a downstream object detection task using generated NIR data to augment limited real NIR datasets. The proposed approach enables the scaling up of NIR datasets without additional data acquisition or annotation efforts, potentially accelerating advancements in NIR-based computer vision applications., Comment: 19 pages,12 figures

The low dynamic range (LDR) of common cameras fails to capture the rich contrast in natural scenes, resulting in loss of color and details in saturated pixels. Reconstructing the high dynamic range (HDR) of luminance present in the scene from single LDR photographs is an important task with many applications in computational photography and realistic display of images. The HDR reconstruction task aims to infer the lost details using the context present in the scene, requiring neural networks to understand high-level geometric and illumination cues. This makes it challenging for data-driven algorithms to generate accurate and high-resolution results. In this work, we introduce a physically-inspired remodeling of the HDR reconstruction problem in the intrinsic domain. The intrinsic model allows us to train separate networks to extend the dynamic range in the shading domain and to recover lost color details in the albedo domain. We show that dividing the problem into two simpler sub-tasks improves performance in a wide variety of photographs., Comment: Accepted for ECCV 2024

Intrinsic image decomposition aims to separate the surface reflectance and the effects from the illumination given a single photograph. Due to the complexity of the problem, most prior works assume a single-color illumination and a Lambertian world, which limits their use in illumination-aware image editing applications. In this work, we separate an input image into its diffuse albedo, colorful diffuse shading, and specular residual components. We arrive at our result by gradually removing first the single-color illumination and then the Lambertian-world assumptions. We show that by dividing the problem into easier sub-problems, in-the-wild colorful diffuse shading estimation can be achieved despite the limited ground-truth datasets. Our extended intrinsic model enables illumination-aware analysis of photographs and can be used for image editing applications such as specularity removal and per-pixel white balancing., Comment: 12 pages, 12 figures. Accepted to SIGGRAPH Asia 2024 (TOG). Webpage: https://yaksoy.github.io/ColorfulShading

Class-agnostic image segmentation is a crucial component in automating image editing workflows, especially in contexts where object selection traditionally involves interactive tools. Existing methods in the literature often adhere to top-down formulations, following the paradigm of class-based approaches, where object detection precedes per-object segmentation. In this work, we present a novel bottom-up formulation for addressing the class-agnostic segmentation problem. We supervise our network directly on the projective sphere of its feature space, employing losses inspired by metric learning literature as well as losses defined in a novel segmentation-space representation. The segmentation results are obtained through a straightforward mean-shift clustering of the estimated features. Our bottom-up formulation exhibits exceptional generalization capability, even when trained on datasets designed for class-based segmentation. We further showcase the effectiveness of our generic approach by addressing the challenging task of cell and nucleus segmentation. We believe that our bottom-up formulation will offer valuable insights into diverse segmentation challenges in the literature.

This paper describes an efficient, open source PyTorch implementation of the Spectral Transform Unit. We investigate sequence prediction tasks over several modalities including language, robotics, and simulated dynamical systems. We find that for the same parameter count, the STU and its variants outperform the Transformer as well as other leading state space models across various modalities.

Bugs are inevitable in software development, and their reporting in open repositories can enhance software transparency and reliability assessment. This study aims to extract information from the issue tracking system Jira and proposes a methodology to estimate resolution time for new bugs. The methodology is applied to network project ONAP, addressing concerns of network operators and manufacturers. This research provides insights into bug resolution times and related aspects in network softwarization projects.

Providing high quality software and evaluating the software reliability in softwarized networks are crucial for vendors and customers. These networks rely on open source code, which are sensitive to contain high number of bugs. Both, the knowledge about the code of previous releases as well as the bug history of the particular project can be used to evaluate the software reliability of a new software release based on SRGM. In this work a framework to predict the number of the bugs of a new release, as well as other reliability parameters, is proposed. An exemplary implementation of this framework to two particular open source projects, is described in detail. The difference between the prediction accuracy of the two projects is presented. Different alternatives to increase the prediction accuracy are proposed and compared in this paper.

Existing methods for scale-invariant monocular depth estimation (SI MDE) often struggle due to the complexity of the task, and limited and non-diverse datasets, hindering generalizability in real-world scenarios. This is while shift-and-scale-invariant (SSI) depth estimation, simplifying the task and enabling training with abundant stereo datasets achieves high performance. We present a novel approach that leverages SSI inputs to enhance SI depth estimation, streamlining the network's role and facilitating in-the-wild generalization for SI depth estimation while only using a synthetic dataset for training. Emphasizing the generation of high-resolution details, we introduce a novel sparse ordinal loss that substantially improves detail generation in SSI MDE, addressing critical limitations in existing approaches. Through in-the-wild qualitative examples and zero-shot evaluation we substantiate the practical utility of our approach in computational photography applications, showcasing its ability to generate highly detailed SI depth maps and achieve generalization in diverse scenarios., Comment: To appear in Proc. SIGGRAPH, 2024. Project webpage: https://yaksoy.github.io/sidepth/

The demand for novel detector mediums such as Water-based Liquid Scintillator (WbLS) has increased over the last few decades due to their capability for both low energy particle interactions and higher light yield. Recently, the usage of machine learning (ML) methods in high-energy physics has also been increasing. The ML and AI methods are used in many physics projects in the field since they provide effective and sensitive results. In this study, we aimed to develop a comprehensive analysis of water Cherenkov detectors and perform physics analyses to efficiently separate Cherenkov and scintillation photons with ML algorithms using the data from the WbLS detector environment. The main goal of this study was to produce more precise solutions to physics problems, such as signal classification, by applying ML techniques to the simulation and experimental data. Here, we trained more than 20 ML models, and our results revealed that three machine learning models, XGBoost, Light GBM, and Random Forest models, and their ensemble model gave us more than 95\% accuracy for separating Cherenkov and scintillation photons with balanced and unbalanced datasets. This is a significant increase in efficiency as compared with the results of the classical method by applying simple time cuts.

In recent years, there has been a growing interest in using Machine Learning (ML), especially Deep Learning (DL) to solve Network Intrusion Detection (NID) problems. However, the feature distribution shift problem remains a difficulty, because the change in features' distributions over time negatively impacts the model's performance. As one promising solution, model pretraining has emerged as a novel training paradigm, which brings robustness against feature distribution shift and has proven to be successful in Computer Vision (CV) and Natural Language Processing (NLP). To verify whether this paradigm is beneficial for NID problem, we propose SwapCon, a ML model in the context of NID, which compresses shift-invariant feature information during the pretraining stage and refines during the finetuning stage. We exemplify the evidence of feature distribution shift using the Kyoto2006+ dataset. We demonstrate how pretraining a model with the proper size can increase robustness against feature distribution shifts by over 8%. Moreover, we show how an adequate numerical embedding strategy also enhances the performance of pretrained models. Further experiments show that the proposed SwapCon model also outperforms eXtreme Gradient Boosting (XGBoost) and K-Nearest Neighbor (KNN) based models by a large margin., Comment: accepted by ICMLCN24

Despite significant advancements in network-based image harmonization techniques, there still exists a domain disparity between typical training pairs and real-world composites encountered during inference. Most existing methods are trained to reverse global edits made on segmented image regions, which fail to accurately capture the lighting inconsistencies between the foreground and background found in composited images. In this work, we introduce a self-supervised illumination harmonization approach formulated in the intrinsic image domain. First, we estimate a simple global lighting model from mid-level vision representations to generate a rough shading for the foreground region. A network then refines this inferred shading to generate a harmonious re-shading that aligns with the background scene. In order to match the color appearance of the foreground and background, we utilize ideas from prior harmonization approaches to perform parameterized image edits in the albedo domain. To validate the effectiveness of our approach, we present results from challenging real-world composites and conduct a user study to objectively measure the enhanced realism achieved compared to state-of-the-art harmonization methods., Comment: 10 pages, 8 figures. Accepted to SIGGRAPH Asia 2023 (Conference Track). Project page: https://yaksoy.github.io/intrinsicCompositing/

Intrinsic decomposition is a fundamental mid-level vision problem that plays a crucial role in various inverse rendering and computational photography pipelines. Generating highly accurate intrinsic decompositions is an inherently under-constrained task that requires precisely estimating continuous-valued shading and albedo. In this work, we achieve high-resolution intrinsic decomposition by breaking the problem into two parts. First, we present a dense ordinal shading formulation using a shift- and scale-invariant loss in order to estimate ordinal shading cues without restricting the predictions to obey the intrinsic model. We then combine low- and high-resolution ordinal estimations using a second network to generate a shading estimate with both global coherency and local details. We encourage the model to learn an accurate decomposition by computing losses on the estimated shading as well as the albedo implied by the intrinsic model. We develop a straightforward method for generating dense pseudo ground truth using our model's predictions and multi-illumination data, enabling generalization to in-the-wild imagery. We present an exhaustive qualitative and quantitative analysis of our predicted intrinsic components against state-of-the-art methods. Finally, we demonstrate the real-world applicability of our estimations by performing otherwise difficult editing tasks such as recoloring and relighting., Comment: 24 pages, 23 figures, Accepted to ACM Transactions on Graphics (2023). Project page: https://yaksoy.github.io/intrinsic/

URLs are fundamental elements of web applications. By applying vector algorithms, we built a fast standard-compliant C++ implementation. Our parser uses three times fewer instructions than competing parsers following the WHATWG standard (e.g., Servo's rust-url) and up to eight times fewer instructions than the popular curl parser. The Node.js environment adopted our C++ library. In our tests on realistic data, a recent Node.js version (20.0) with our parser is four to five times faster than the last version with the legacy URL parser.

Common editing operations performed by professional photographers include the cleanup operations: de-emphasizing distracting elements and enhancing subjects. These edits are challenging, requiring a delicate balance between manipulating the viewer's attention while maintaining photo realism. While recent approaches can boast successful examples of attention attenuation or amplification, most of them also suffer from frequent unrealistic edits. We propose a realism loss for saliency-guided image enhancement to maintain high realism across varying image types, while attenuating distractors and amplifying objects of interest. Evaluations with professional photographers confirm that we achieve the dual objective of realism and effectiveness, and outperform the recent approaches on their own datasets, while requiring a smaller memory footprint and runtime. We thus offer a viable solution for automating image enhancement and photo cleanup operations., Comment: For more info visit http://yaksoy.github.io/realisticEditing/

Flash is an essential tool as it often serves as the sole controllable light source in everyday photography. However, the use of flash is a binary decision at the time a photograph is captured with limited control over its characteristics such as strength or color. In this work, we study the computational control of the flash light in photographs taken with or without flash. We present a physically motivated intrinsic formulation for flash photograph formation and develop flash decomposition and generation methods for flash and no-flash photographs, respectively. We demonstrate that our intrinsic formulation outperforms alternatives in the literature and allows us to computationally control flash in in-the-wild images., Comment: 9 pages, 15 figures. Accepted to CVPR 2023. Project page: http://yaksoy.github.io/intrinsicFlash/

Quantum emitters in solid-state crystals have recently attracted a lot of attention due to their simple applicability in optical quantum technologies. The polarization of single photons generated by quantum emitters is one of the key parameters that play a crucial role in the applications, such as quantum computation that uses the indistinguishability of photons. However, the degree of single photon polarization is typically quantified using time-averaged photoluminescence intensity of single emitters, which provides limited information about the dipole properties in solids. In this work, we use single defects in hexagonal boron nitride and nanodiamond as efficient room-temperature single photon sources to reveal the origin and the temporal evolution of dipole orientation in solid-state quantum emitters. The angle of excitation and emission dipoles relative to the crystal axes are determined experimentally and then calculated using density functional theory, which results in characteristic angles for every specific defect that can be used as an efficient tool for defect identification and understanding their atomic structure. Moreover, the temporal polarization dynamics reveal a strongly modified linear polarization visibility that depends on the excited state decay time of individual excitation. This effect can be traced back potentially to the excitation of excess charges in the local crystal environment. Understanding such hidden time-dependent mechanisms can further be used to improve the performance of polarization-sensitive experiments, in particular that of quantum communication with single photon emitters.

Semantic signal processing and communications are poised to play a central part in developing the next generation of sensor devices and networks. A crucial component of a semantic system is the extraction of semantic signals from the raw input signals, which has become increasingly tractable with the recent advances in machine learning (ML) and artificial intelligence (AI) techniques. The accurate extraction of semantic signals using the aforementioned ML and AI methods, and the detection of semantic innovation for scheduling transmission and/or storage events are critical tasks for reliable semantic signal processing and communications. In this work, we propose a reliable semantic information extraction framework based on our previous work on semantic signal representations in a hierarchical graph-based structure. The proposed framework includes a time integration method to increase fidelity of ML outputs in a class-aware manner, a graph-edit-distance based metric to detect innovation events at the graph-level and filter out sporadic errors, and a Hidden Markov Model (HMM) to produce smooth and reliable graph signals. The proposed methods within the framework are demonstrated individually and collectively through simulations and case studies based on real-world computer vision examples.