Search

Community vision screening plays a crucial role in identifying individuals with vision loss and preventing avoidable blindness, particularly in rural communities where access to eye care services is limited. Currently, there is a pressing need for a simple and efficient process to screen and refer individuals with significant eye disease-related vision loss to tertiary eye care centers for further care. An ideal solution should seamlessly and readily integrate with existing workflows, providing comprehensive initial screening results to service providers, thereby enabling precise patient referrals for timely treatment. This paper introduces the Enhancing Community Vision Screening (ECVS) solution, which addresses the aforementioned concerns with a novel and feasible solution based on simple, non-invasive retinal photography for the detection of pathology-based visual impairment. Our study employs four distinct deep learning models: RETinal photo Quality Assessment (RETQA), Pathology Visual Impairment detection (PVI), Eye Disease Diagnosis (EDD) and Visualization of Lesion Regions of the eye (VLR). We conducted experiments on over 10 datasets, totaling more than 80,000 fundus photos collected from various sources. The models integrated into ECVS achieved impressive AUC scores of 0.98 for RETQA, 0.95 for PVI, and 0.90 for EDD, along with a DICE coefficient of 0.48 for VLR. These results underscore the promising capabilities of ECVS as a straightforward and scalable method for community-based vision screening., Comment: 11 pages, 4 figures, published in MICCAI2024 OMIA XI workshop

Current metadata creation for web archives is time consuming and costly due to reliance on human effort. This paper explores the use of gpt-4o for metadata generation within the Web Archive Singapore, focusing on scalability, efficiency, and cost effectiveness. We processed 112 Web ARChive (WARC) files using data reduction techniques, achieving a notable 99.9% reduction in metadata generation costs. By prompt engineering, we generated titles and abstracts, which were evaluated both intrinsically using Levenshtein Distance and BERTScore, and extrinsically with human cataloguers using McNemar's test. Results indicate that while our method offers significant cost savings and efficiency gains, human curated metadata maintains an edge in quality. The study identifies key challenges including content inaccuracies, hallucinations, and translation issues, suggesting that Large Language Models (LLMs) should serve as complements rather than replacements for human cataloguers. Future work will focus on refining prompts, improving content filtering, and addressing privacy concerns through experimentation with smaller models. This research advances the integration of LLMs in web archiving, offering valuable insights into their current capabilities and outlining directions for future enhancements. The code is available at https://github.com/masamune-prog/warc2summary for further development and use by institutions facing similar challenges.

We investigate whether Neural Networks (NN) can accurately differentiate between growth-rate data of the Large Scale Structure (LSS) of the Universe, simulated via two models: a cosmological constant and cold dark matter ({\Lambda}CDM) model and a tomographic Coupled Dark Energy (CDE) model. We build an NN classifier and test its accuracy in distinguishing between cosmological models. For our dataset, we generate f{\sigma_8}(z) growth-rate observables simulating a realistic Stage IV galaxy survey-like setup, for both {\Lambda}CDM and a tomographic CDE model, for various values of the model parameters. We then optimise and train our NN with Optuna, aiming to avoid overfitting and maximising the accuracy of the trained model. We conduct our analysis for both a binary classification, comparing between {\Lambda}CDM and a CDE model where only one tomographic coupling bin is activated, and a multiclass classification scenario where all the models are combined. For the case of binary classification, we find that our NN can confidently (with > 86% accuracy) detect non-zero values of the tomographic coupling regardless of the redshift range at which coupling is activated, and at a 100% confidence level, detect the {\Lambda}CDM model. For the multiclass classification task, we find that the NN performs adequately well at distinguishing between {\Lambda}CDM, a CDE model with low redshift coupling, and a model with high redshift coupling, with 99%, 79% and 84% accuracy respectively. By leveraging the power of machine learning, our pipeline can be a useful tool to analyse growth-rate data and maximise the potential of current surveys to probe for deviations from General Relativity., Comment: Accepted for publication in A&A

Medical Vision-Language Pretraining (MedVLP) shows promise in learning generalizable and transferable visual representations from paired and unpaired medical images and reports. MedVLP can provide useful features to downstream tasks and facilitate adapting task-specific models to new setups using fewer examples. However, existing MedVLP methods often differ in terms of datasets, preprocessing, and finetuning implementations. This pose great challenges in evaluating how well a MedVLP method generalizes to various clinically-relevant tasks due to the lack of unified, standardized, and comprehensive benchmark. To fill this gap, we propose BenchX, a unified benchmark framework that enables head-to-head comparison and systematical analysis between MedVLP methods using public chest X-ray datasets. Specifically, BenchX is composed of three components: 1) Comprehensive datasets covering nine datasets and four medical tasks; 2) Benchmark suites to standardize data preprocessing, train-test splits, and parameter selection; 3) Unified finetuning protocols that accommodate heterogeneous MedVLP methods for consistent task adaptation in classification, segmentation, and report generation, respectively. Utilizing BenchX, we establish baselines for nine state-of-the-art MedVLP methods and found that the performance of some early MedVLP methods can be enhanced to surpass more recent ones, prompting a revisiting of the developments and conclusions from prior works in MedVLP. Our code are available at https://github.com/yangzhou12/BenchX., Comment: Accepted to NeurIPS24 Datasets and Benchmarks Track

Traditional blockchain systems, such as Ethereum, typically rely on a \emph{single volatile cryptocurrency for transaction fees}. This leads to fluctuating transaction fee prices and limits the flexibility of users' payment options. To address these issues, we propose MStableChain, which leverage multiple stablecoins as native tokens for transaction fee settlements, thus ensuring stable transaction fees and flexible payment options. To address the challenges of mass adoption and practicality, we propose several core designs. To maintain compatibility with the Ethereum Virtual Machine (EVM) for mass adoption while supporting multiple native stablecoins, MStableChain employs a multi-currency units, multi-type RPCs mechanism. This mechanism enables the system to handle multiple stablecoins without altering the EVM or requiring changes to user applications. Furthermore, an oracle-based gas fee adjustment mechanism is proposed to manage exchange rates between different stablecoins, ensuring equitable transaction costs across various currencies. The system also introduces a secure, on-chain voting-based management protocol for the administrative functions related to these stablecoins. Experimental results from a prototype implementation demonstrate that MStableChain provides stable transaction fee prices, high effectiveness, and good usability., Comment: In submission to IEEE TSC

Deep learning-based expert models have reached superhuman performance in decision-making domains such as chess and Go. However, it is under-explored to explain or comment on given decisions although it is important for human education and model explainability. The outputs of expert models are accurate, but yet difficult to interpret for humans. On the other hand, large language models (LLMs) produce fluent commentary but are prone to hallucinations due to their limited decision-making capabilities. To bridge this gap between expert models and LLMs, we focus on chess commentary as a representative case of explaining complex decision-making processes through language and address both the generation and evaluation of commentary. We introduce Concept-guided Chess Commentary generation (CCC) for producing commentary and GPT-based Chess Commentary Evaluation (GCC-Eval) for assessing it. CCC integrates the decision-making strengths of expert models with the linguistic fluency of LLMs through prioritized, concept-based explanations. GCC-Eval leverages expert knowledge to evaluate chess commentary based on informativeness and linguistic quality. Experimental results, validated by both human judges and GCC-Eval, demonstrate that CCC generates commentary that is accurate, informative, and fluent.

GPT-4o is an autoregressive omni model that accepts as input any combination of text, audio, image, and video, and generates any combination of text, audio, and image outputs. It's trained end-to-end across text, vision, and audio, meaning all inputs and outputs are processed by the same neural network. GPT-4o can respond to audio inputs in as little as 232 milliseconds, with an average of 320 milliseconds, which is similar to human response time in conversation. It matches GPT-4 Turbo performance on text in English and code, with significant improvement on text in non-English languages, while also being much faster and 50\% cheaper in the API. GPT-4o is especially better at vision and audio understanding compared to existing models. In line with our commitment to building AI safely and consistent with our voluntary commitments to the White House, we are sharing the GPT-4o System Card, which includes our Preparedness Framework evaluations. In this System Card, we provide a detailed look at GPT-4o's capabilities, limitations, and safety evaluations across multiple categories, focusing on speech-to-speech while also evaluating text and image capabilities, and measures we've implemented to ensure the model is safe and aligned. We also include third-party assessments on dangerous capabilities, as well as discussion of potential societal impacts of GPT-4o's text and vision capabilities.

Workplace safety has received increasing attention as millions of workers worldwide suffer from work-related accidents. Despite poor housekeeping is a significant contributor to construction accidents, there remains a significant lack of technological research focused on improving housekeeping practices in construction sites. Recognizing and locating poor housekeeping in a dynamic construction site is an important task that can be improved through computer vision approaches. Despite advances in AI and computer vision, existing methods for detecting poor housekeeping conditions face many challenges, including limited explanations, lack of locating of poor housekeeping, and lack of annotated datasets. On the other hand, change detection which aims to detect the changed environmental conditions (e.g., changing from good to poor housekeeping) and 'where' the change has occurred (e.g., location of objects causing poor housekeeping), has not been explored to the problem of housekeeping management. To address these challenges, we propose the Housekeeping Change Detection Network (HCDN), an advanced change detection neural network that integrates a feature fusion module and a large vision model, achieving state-of-the-art performance. Additionally, we introduce the approach to establish a novel change detection dataset (named Housekeeping-CCD) focused on housekeeping in construction sites, along with a housekeeping segmentation dataset. Our contributions include significant performance improvements compared to existing methods, providing an effective tool for enhancing construction housekeeping and safety. To promote further development, we share our source code and trained models for global researchers: https://github.com/NUS-DBE/Housekeeping-CD.

Despite the success of Large Language Models (LLMs) across various fields, their potential to generate untruthful, biased and harmful responses poses significant risks, particularly in critical applications. This highlights the urgent need for systematic methods to detect and prevent such misbehavior. While existing approaches target specific issues such as harmful responses, this work introduces LLMScan, an innovative LLM monitoring technique based on causality analysis, offering a comprehensive solution. LLMScan systematically monitors the inner workings of an LLM through the lens of causal inference, operating on the premise that the LLM's `brain' behaves differently when misbehaving. By analyzing the causal contributions of the LLM's input tokens and transformer layers, LLMScan effectively detects misbehavior. Extensive experiments across various tasks and models reveal clear distinctions in the causal distributions between normal behavior and misbehavior, enabling the development of accurate, lightweight detectors for a variety of misbehavior detection tasks.

Large language models (LLMs) have garnered substantial attention due to their promising applications in diverse domains. Nevertheless, the increasing size of LLMs comes with a significant surge in the computational requirements for training and deployment. Memristor crossbars have emerged as a promising solution, which demonstrated a small footprint and remarkably high energy efficiency in computer vision (CV) models. Memristors possess higher density compared to conventional memory technologies, making them highly suitable for effectively managing the extreme model size associated with LLMs. However, deploying LLMs on memristor crossbars faces three major challenges. Firstly, the size of LLMs increases rapidly, already surpassing the capabilities of state-of-the-art memristor chips. Secondly, LLMs often incorporate multi-head attention blocks, which involve non-weight stationary multiplications that traditional memristor crossbars cannot support. Third, while memristor crossbars excel at performing linear operations, they are not capable of executing complex nonlinear operations in LLM such as softmax and layer normalization. To address these challenges, we present a novel architecture for the memristor crossbar that enables the deployment of state-of-the-art LLM on a single chip or package, eliminating the energy and time inefficiencies associated with off-chip communication. Our testing on BERT_Large showed negligible accuracy loss. Compared to traditional memristor crossbars, our architecture achieves enhancements of up to 39X in area overhead and 18X in energy consumption. Compared to modern TPU/GPU systems, our architecture demonstrates at least a 68X reduction in the area-delay product and a significant 69% energy consumption reduction.

Colorectal cancer is the most common cancer in Singapore and the third most common cancer worldwide. Blood in a person's stool is a symptom of this disease, and it is usually detected by the faecal occult blood test (FOBT). However, the FOBT presents several limitations - the collection process for the stool samples is tedious and unpleasant, the waiting period for results is about 2 weeks and costs are involved. In this research, we propose a simple-to-use, fast and cost-free alternative - a stool recognition neural network that determines if there is blood in one's stool (which indicates a possible risk of colorectal cancer) from an image of it. As this is a new classification task, there was limited data available, hindering classifier performance. Hence, various Generative Adversarial Networks (GANs) (DiffAugment StyleGAN2, DCGAN, Conditional GAN) were trained to generate images of high fidelity to supplement the dataset. Subsequently, images generated by the GAN with the most realistic images (DiffAugment StyleGAN2) were concatenated to the classifier's training batch on-the-fly, improving accuracy to 94%. This model was then deployed to a mobile app - Poolice, where users can take a photo of their stool and obtain instantaneous results if there is blood in their stool, prompting those who do to seek medical advice. As "early detection saves lives", we hope our app built on our stool recognition neural network can help people detect colorectal cancer earlier, so they can seek treatment and have higher chances of survival., Comment: 21 pages, 28 figures

Developing strategies to systematically increase the critical current, the threshold current below which the superconductivity exists, is an important goal of materials science. Here, the concept of quantum phase transition is employed to enhance the critical current of a kagome superconductor CsV$_3$Sb$_5$, which exhibits a charge density wave (CDW) and superconductivity that are both affected by hydrostatic pressure. As the CDW phase is rapidly suppressed under pressure, a large enhancement in the self-field critical current ($I_{\rm c,sf}$) is recorded. The observation of a peak-like enhancement of $I_{\rm c,sf}$ at the zero-temperature limit ($I_{\rm c,sf}(0)$) centred at $p^*\approx 20$~kbar, the same pressure where the CDW phase transition vanishes, further provides strong evidence of a zero-temperature quantum anomaly in this class of pressure-tuned superconductor. Such a peak in $I_{\rm c,sf}(0)$ resembles the findings in other well-established quantum-critical superconductors, hinting at the presence of enhanced quantum fluctuations associated with the CDW phase in CsV$_3$Sb$_5$., Comment: 8 pages, 4 figures. Advanced Science (2024)

Improving the fairness of federated learning (FL) benefits healthy and sustainable collaboration, especially for medical applications. However, existing fair FL methods ignore the specific characteristics of medical FL applications, i.e., domain shift among the datasets from different hospitals. In this work, we propose Fed-LWR to improve performance fairness from the perspective of feature shift, a key issue influencing the performance of medical FL systems caused by domain shift. Specifically, we dynamically perceive the bias of the global model across all hospitals by estimating the layer-wise difference in feature representations between local and global models. To minimize global divergence, we assign higher weights to hospitals with larger differences. The estimated client weights help us to re-aggregate the local models per layer to obtain a fairer global model. We evaluate our method on two widely used federated medical image segmentation benchmarks. The results demonstrate that our method achieves better and fairer performance compared with several state-of-the-art fair FL methods., Comment: 11 pages, 2 Figures

Large vision language models (VLMs) combine large language models with vision encoders, demonstrating promise across various tasks. However, they often underperform in task-specific applications due to domain gaps between pre-training and fine-tuning. We introduce VITask, a novel framework that enhances task-specific adaptability of VLMs by integrating task-specific models (TSMs). VITask employs three key strategies: exemplar prompting (EP), response distribution alignment (RDA), and contrastive response tuning (CRT) to improve the task-specific performance of VLMs by adjusting their response distributions. EP allows TSM features to guide VLMs, while RDA enables VLMs to adapt without TSMs during inference by learning from exemplar-prompted models. CRT further optimizes the ranking of correct image-response pairs, thereby reducing the risk of generating undesired responses. Experiments on 12 medical diagnosis datasets across 9 imaging modalities show that VITask outperforms both vanilla instruction-tuned VLMs and TSMs, showcasing its ability to integrate complementary features from both models effectively. Additionally, VITask offers practical advantages such as flexible TSM integration and robustness to incomplete instructions, making it a versatile and efficient solution for task-specific VLM tuning. Our code are available at https://github.com/baiyang4/VITask.

While fine-tuning pretrained models has become common practice, these models often underperform outside their specific domains. Recently developed model merging techniques enable the direct integration of multiple models, each fine-tuned for distinct tasks, into a single model. This strategy promotes multitasking capabilities without requiring retraining on the original datasets. However, existing methods fall short in addressing potential conflicts and complex correlations between tasks, especially in parameter-level adjustments, posing a challenge in effectively balancing parameter competition across various tasks. This paper introduces an innovative technique named PCB-Merging (Parameter Competition Balancing), a lightweight and training-free technique that adjusts the coefficients of each parameter for effective model merging. PCB-Merging employs intra-balancing to gauge parameter significance within individual tasks and inter-balancing to assess parameter similarities across different tasks. Parameters with low importance scores are dropped, and the remaining ones are rescaled to form the final merged model. We assessed our approach in diverse merging scenarios, including cross-task, cross-domain, and cross-training configurations, as well as out-of-domain generalization. The experimental results reveal that our approach achieves substantial performance enhancements across multiple modalities, domains, model sizes, number of tasks, fine-tuning forms, and large language models, outperforming existing model merging methods. The code is publicly available at: \url{https://github.com/duguodong7/pcb-merging}., Comment: Accepted by NeurIPS2024

Quantum sensing is one of the most promising applications for quantum technologies. However, reaching the ultimate sensitivities enabled by the laws of quantum mechanics can be a challenging task in realistic scenarios where noise is present. While several strategies have been proposed to deal with the detrimental effects of noise, these come at the cost of an extra shot budget. Given that shots are a precious resource for sensing -- as infinite measurements could lead to infinite precision -- care must be taken to truly guarantee that any shot not being used for sensing is actually leading to some metrological improvement. In this work, we study whether investing shots in error-mitigation, inference techniques, or combinations thereof, can improve the sensitivity of a noisy quantum sensor on a (shot) budget. We present a detailed bias-variance error analysis for various sensing protocols. Our results show that the costs of zero-noise extrapolation techniques outweigh their benefits. We also find that pre-characterizing a quantum sensor via inference techniques leads to the best performance, under the assumption that the sensor is sufficiently stable., Comment: 17+18 pages, 8+2 figures, 1+1 tables

When observing the chip-to-free-space light beams from silicon photonics (SiPh) to free-space, manual adjustment of camera lens is often required to obtain a focused image of the light beams. In this letter, we demonstrated an auto-focusing system based on you-only-look-once (YOLO) model. The trained YOLO model exhibits high classification accuracy of 99.7% and high confidence level >0.95 when detecting light beams from SiPh gratings. A video demonstration of real-time light beam detection, real-time computation of beam width, and auto focusing of light beams are also included.

The ribbon number of a knot is the minimum number of ribbon singularities among all ribbon disks bounded by that knot. In this paper, we build on the systematic treatment of this knot invariant initiated in recent work of Friedl, Misev, and Zupan. We show that the set of Alexander polynomials of knots with ribbon number at most four contains 56 polynomials, and we use this set to compute the ribbon numbers for many 12-crossing knots. We also study higher-genus ribbon numbers of knots, presenting some examples that exhibit interesting behavior and establishing that the success of the Alexander polynomial at controlling genus-0 ribbon numbers does not extend to higher genera., Comment: 32 pages, 12 figures, 7 tables

Inertial microfluidics has been limited to dilute particle concentrations due to defocusing (spreading out) at high particle concentrations. We observe a counterintuitive shift of focusing to the outer curved wall under high concentration flow, which contradicts the existing particle focusing theory. We developed a multiphase model incorporating lift forces and particle-particle interactions to explain this behaviour. Numerical simulations validated by experimental data reveal the shift is governed by the ratio of the lift force strength to that of particle interaction frequencies.

Two-dimensional van der Waals semiconductors are promising for future nanoelectronics. However, integrating high-k gate dielectrics for device applications is challenging as the inert van der Waals material surfaces hinder uniform dielectric growth. Here, we report a liquid metal oxide-assisted approach to integrate ultrathin, high-k HfO2 dielectric on 2D semiconductors with atomically smooth interfaces. Using this approach, we fabricated 2D WS2 top-gated transistors with subthreshold swings down to 74.5 mV/dec, gate leakage current density below 10-6 A/cm2, and negligible hysteresis. We further demonstrate a one-step van der Waals integration of contacts and dielectrics on graphene. This can offer a scalable approach toward integrating entire prefabricated device stack arrays with 2D materials. Our work provides a scalable solution to address the crucial dielectric engineering challenge for 2D semiconductors, paving the way for high-performance 2D electronics.

Future electronics require aggressive scaling of channel material thickness while maintaining device performance. Two-dimensional (2D) semiconductors are promising candidates, but despite over two decades of research, experimental performance still lags theoretical expectations. Here, we develop an oxygen-free approach to push the electrical transport of 2D field-effect transistors toward the theoretical phonon-limited intrinsic mobility. We achieve record carrier mobilities of 91 (132) cm2V-1s-1 for mono- (bi-) layer MoS2 transistors on SiO2 substrate. Statistics from over 60 devices confirm that oxygen-free fabrication enhances key figures of merit by more than an order of magnitude. While previous studies suggest that 2D transition metal dichalcogenides such as MoS2 and WS2 are stable in air, we show that short-term ambient exposure can degrade their device performance through irreversible oxygen chemisorption. This study emphasizes the criticality of avoiding oxygen exposure, offering guidance for device manufacturing for fundamental research and practical applications of 2D materials.

The maritime industry stands at a critical juncture, where the imperative for technological advancement intersects with the pressing need for robust cybersecurity measures. Maritime cybersecurity refers to the protection of computer systems and digital assests within the maritime industry, as well as the broader network of interconnected components that make up the maritime ecosystem. In this survey, we aim to identify the significant domains of maritime cybersecurity and measure their effectiveness. We have provided an in-depth analysis of threats in key maritime systems, including AIS, GNSS, ECDIS, VDR, RADAR, VSAT, and GMDSS, while exploring real-world cyber incidents that have impacted the sector. A multi-dimensional taxonomy of maritime cyber attacks is presented, offering insights into threat actors, motivations, and impacts. We have also evaluated various security solutions, from integrated solutions to component specific solutions. Finally, we have shared open challenges and future solutions. In the supplementary section, we have presented definitions and vulnerabilities of vessel components that have discussed in this survey. By addressing all these critical issues with key interconnected aspects, this review aims to foster a more resilient maritime ecosystem., Comment: 23 pages, long survey paper, submitted to IEEE journals because ACM computing survey is not a good fit in terms of their dedicated scope

The volunteer's dilemma is a well-known game in game theory that models the conflict players face when deciding whether to volunteer for a collective benefit, knowing that volunteering incurs a personal cost. In this work, we introduce a quantum variant of the classical volunteer's dilemma, generalizing it by allowing players to utilize quantum strategies. Employing the Eisert-Wilkens-Lewenstein quantization framework, we analyze a multiplayer quantum volunteer's dilemma scenario with an arbitrary number of players, where the cost of volunteering is shared equally among the volunteers. We derive analytical expressions for the players' expected payoffs and demonstrate the quantum game's advantage over the classical game. In particular, we prove that the quantum volunteer's dilemma possesses symmetric Nash equilibria with larger expected payoffs compared to the unique symmetric Nash equilibrium of the classical game, wherein players use mixed strategies. Furthermore, we show that the quantum Nash equilibria we identify are Pareto optimal. Our findings reveal distinct dynamics in volunteer's dilemma scenarios when players adhere to quantum rules, underscoring a strategic advantage of decision-making in quantum settings., Comment: 28 pages, 5 figures

We present the first measurement of the missing energy due to nuclear effects in monoenergetic, muon neutrino charged-current interactions on carbon, originating from $K^+ \rightarrow \mu^+ \nu_\mu$ decay-at-rest ($E_{\nu_\mu}=235.5$ MeV), performed with the JSNS$^2$ liquid scintillator based experiment. Towards characterizing the neutrino interaction, ostensibly $\nu_\mu n \rightarrow \mu^- p$ or $\nu_\mu$$^{12}\mathrm{C}$ $\rightarrow \mu^-$$^{12}\mathrm{N}$, and in analogy to similar electron scattering based measurements, we define the missing energy as the energy transferred to the nucleus ($\omega$) minus the kinetic energy of the outgoing proton(s), $E_{m} \equiv \omega-\sum T_p$, and relate this to visible energy in the detector, $E_{m}=E_{\nu_\mu}~(235.5~\mathrm{MeV})-m_\mu~(105.7~\mathrm{MeV}) - E_{vis}$. The missing energy, which is naively expected to be zero in the absence of nuclear effects (e.g. nucleon separation energy, Fermi momenta, and final-state interactions), is uniquely sensitive to many aspects of the interaction, and has previously been inaccessible with neutrinos. The shape-only, differential cross section measurement reported, based on a $(77\pm3)$% pure double-coincidence KDAR signal (621 total events), provides an important benchmark for models and event generators at 100s-of-MeV neutrino energies, characterized by the difficult-to-model transition region between neutrino-nucleus and neutrino-nucleon scattering, and relevant for applications in nuclear physics, neutrino oscillation measurements, and Type-II supernova studies.

AI Safety is an emerging area of critical importance to the safe adoption and deployment of AI systems. With the rapid proliferation of AI and especially with the recent advancement of Generative AI (or GAI), the technology ecosystem behind the design, development, adoption, and deployment of AI systems has drastically changed, broadening the scope of AI Safety to address impacts on public safety and national security. In this paper, we propose a novel architectural framework for understanding and analyzing AI Safety; defining its characteristics from three perspectives: Trustworthy AI, Responsible AI, and Safe AI. We provide an extensive review of current research and advancements in AI safety from these perspectives, highlighting their key challenges and mitigation approaches. Through examples from state-of-the-art technologies, particularly Large Language Models (LLMs), we present innovative mechanism, methodologies, and techniques for designing and testing AI safety. Our goal is to promote advancement in AI safety research, and ultimately enhance people's trust in digital transformation.

Over the past decade, there has been extensive work in developing integrated silicon photonics (SiPh) gratings for the optical addressing of trapped ion qubits in the ion trap quantum computing community. However, when viewing beam profiles from infrared (IR) cameras, it is often difficult to determine the corresponding heights where the beam profiles are located. In this work, we developed transformer models to recognize the corresponding height categories of beam profiles of light from SiPh gratings. The model is trained using two techniques: (1) input patches, and (2) input sequence. For model trained with input patches, the model achieved recognition accuracy of 0.938. Meanwhile, model trained with input sequence shows lower accuracy of 0.895. However, when repeating the model-training 150 cycles, model trained with input patches shows inconsistent accuracy ranges between 0.445 to 0.959, while model trained with input sequence exhibit higher accuracy values between 0.789 to 0.936. The obtained outcomes can be expanded to various applications, including auto-focusing of light beam and auto-adjustment of z-axis stage to acquire desired beam profiles.

This paper explores the use of Grover's Algorithm in the classical rainbow table, uncovering the potential of integrating quantum computing techniques with conventional cryptographic methods to develop a Quantum Rainbow Table Proof-of-Concept. This leverages on Quantum concepts and algorithms which includes the principle of qubit superposition, entanglement and teleportation, coupled with Grover's Algorithm to enable a more efficient search through the rainbow table. The paper also details on the hardware constraints and the work around to produce better results in the implementation stages. Through this work we develop a working prototype of quantum rainbow table and demonstrate how quantum computing could significantly improve the speed of cyber tools such as password crackers and thus impact the cyber security landscape.

There is growing interest in using safety analytics and machine learning to support the prevention of workplace incidents, especially in high-risk industries like construction and trucking. Although existing safety analytics studies have made remarkable progress, they suffer from imbalanced datasets, a common problem in safety analytics, resulting in prediction inaccuracies. This can lead to management problems, e.g., incorrect resource allocation and improper interventions. To overcome the imbalanced data problem, we extend the theory of accident triangle to claim that the importance of data samples should be based on characteristics such as injury severity, accident frequency, and accident type. Thus, three oversampling methods are proposed based on assigning different weights to samples in the minority class. We find robust improvements among different machine learning algorithms. For the lack of open-source safety datasets, we are sharing three imbalanced datasets, e.g., a 9-year nationwide construction accident record dataset, and their corresponding codes.

Retinal foundation models aim to learn generalizable representations from diverse retinal images, facilitating label-efficient model adaptation across various ophthalmic tasks. Despite their success, current retinal foundation models are generally restricted to a single imaging modality, such as Color Fundus Photography (CFP) or Optical Coherence Tomography (OCT), limiting their versatility. Moreover, these models may struggle to fully leverage expert annotations and overlook the valuable domain knowledge essential for domain-specific representation learning. To overcome these limitations, we introduce UrFound, a retinal foundation model designed to learn universal representations from both multimodal retinal images and domain knowledge. UrFound is equipped with a modality-agnostic image encoder and accepts either CFP or OCT images as inputs. To integrate domain knowledge into representation learning, we encode expert annotation in text supervision and propose a knowledge-guided masked modeling strategy for model pre-training. It involves reconstructing randomly masked patches of retinal images while predicting masked text tokens conditioned on the corresponding retinal image. This approach aligns multimodal images and textual expert annotations within a unified latent space, facilitating generalizable and domain-specific representation learning. Experimental results demonstrate that UrFound exhibits strong generalization ability and data efficiency when adapting to various tasks in retinal image analysis. By training on ~180k retinal images, UrFound significantly outperforms the state-of-the-art retinal foundation model trained on up to 1.6 million unlabelled images across 8 public retinal datasets. Our code and data are available at https://github.com/yukkai/UrFound.

Drawing inspiration from the philosophy of Yi Jing, the Yin-Yang pair optimization (YYPO) algorithm has been shown to achieve competitive performance in single objective optimizations, in addition to the advantage of low time complexity when compared to other population-based meta-heuristics. Building upon a reversal concept in Yi Jing, we propose the novel Yi optimization (YI) algorithm. Specifically, we enhance the Yin-Yang pair in YYPO with a proposed Yi-point, in which we use Cauchy flight to update the solution, by implementing both the harmony and reversal concept of Yi Jing. The proposed Yi-point balances both the effort of exploration and exploitation in the optimization process. To examine YI, we use the IEEE CEC 2017 benchmarks and compare YI against the dynamical YYPO, CV1.0 optimizer, and four classical optimizers, i.e., the differential evolution, the genetic algorithm, the particle swarm optimization, and the simulated annealing. According to the experimental results, YI shows highly competitive performance while keeping the low time complexity. The results of this work have implications for enhancing a meta-heuristic optimizer using the philosophy of Yi Jing. While this work implements only certain aspects of Yi Jing, we envisage enhanced performance by incorporating other aspects., Comment: This work has been submitted to the IEEE for possible publication. arXiv admin note: substantial text overlap with arXiv:2104.08564

Darwinian evolution of the biological brain is documented through multiple lines of evidence, although the modes of evolutionary changes remain unclear. Drawing inspiration from the evolved neural systems (e.g., visual cortex), deep learning models have demonstrated superior performance in visual tasks, among others. While the success of training deep neural networks has been relying on back-propagation (BP) and its variants to learn representations from data, BP does not incorporate the evolutionary processes that govern biological neural systems. This work proposes a neural network optimization framework based on evolutionary theory. Specifically, BP-trained deep neural networks for visual recognition tasks obtained from the ending epochs are considered the primordial ancestors (initial population). Subsequently, the population evolved with differential evolution. Extensive experiments are carried out to examine the relationships between Darwinian evolution and neural network optimization, including the correspondence between datasets, environment, models, and living species. The empirical results show that the proposed framework has positive impacts on the network, with reduced over-fitting and an order of magnitude lower time complexity compared to BP. Moreover, the experiments show that the proposed framework performs well on deep neural networks and big datasets., Comment: This work has been submitted to the IEEE for possible publication

Neural architecture search (NAS) automates neural network design by using optimization algorithms to navigate architecture spaces, reducing the burden of manual architecture design. While NAS has achieved success, applying it to emerging domains, such as analyzing unstructured 3D point clouds, remains underexplored due to the data lying in non-Euclidean spaces, unlike images. This paper presents Success-History-based Self-adaptive Differential Evolution with a Joint Point Interaction Dimension Search (SHSADE-PIDS), an evolutionary NAS framework that encodes discrete deep neural network architectures to continuous spaces and performs searches in the continuous spaces for efficient point cloud neural architectures. Comprehensive experiments on challenging 3D segmentation and classification benchmarks demonstrate SHSADE-PIDS's capabilities. It discovered highly efficient architectures with higher accuracy, significantly advancing prior NAS techniques. For segmentation on SemanticKITTI, SHSADE-PIDS attained 64.51% mean IoU using only 0.55M parameters and 4.5GMACs, reducing overhead by over 22-26X versus other top methods. For ModelNet40 classification, it achieved 93.4% accuracy with just 1.31M parameters, surpassing larger models. SHSADE-PIDS provided valuable insights into bridging evolutionary algorithms with neural architecture optimization, particularly for emerging frontiers like point cloud learning., Comment: This work has been submitted to the IEEE for possible publication

Existing neural constructive solvers for routing problems have predominantly employed transformer architectures, conceptualizing the route construction as a set-to-sequence learning task. However, their efficacy has primarily been demonstrated on entirely random problem instances that inadequately capture real-world scenarios. In this paper, we introduce realistic Traveling Salesman Problem (TSP) scenarios relevant to industrial settings and derive the following insights: (1) The optimal next node (or city) to visit often lies within proximity to the current node, suggesting the potential benefits of biasing choices based on current locations. (2) Effectively solving the TSP requires robust tracking of unvisited nodes and warrants succinct grouping strategies. Building upon these insights, we propose integrating a learnable choice layer inspired by Hypernetworks to prioritize choices based on the current location, and a learnable approximate clustering algorithm inspired by the Expectation-Maximization algorithm to facilitate grouping the unvisited cities. Together, these two contributions form a hierarchical approach towards solving the realistic TSP by considering both immediate local neighbourhoods and learning an intermediate set of node representations. Our hierarchical approach yields superior performance compared to both classical and recent transformer models, showcasing the efficacy of the key designs., Comment: Accepted to KDD 2024

Ground state preparation is a key area where quantum computers are expected to prove advantageous. Double-bracket quantum algorithms (DBQAs) have been recently proposed to diagonalize Hamiltonians and in this work we show how to use them to prepare ground states. We propose to improve an initial state preparation by adding a few steps of DBQAs. The interfaced method systematically achieves a better fidelity while significantly reducing the computational cost of the procedure. For a Heisenberg model, we compile our algorithm using CZ and single-qubit gates into circuits that match capabilities of near-term quantum devices. Moreover, we show that DBQAs can benefit from the experimental availability of increasing circuit depths. Whenever an approximate ground state can be prepared without exhausting the available circuit depth, then DBQAs can be enlisted to algorithmically seek a higher fidelity preparation., Comment: 5 pages + appendix, 4 figures, code available at: https://github.com/qiboteam/boostvqe

Parkinson's Disease (PD) diagnosis remains challenging. This study applies Convolutional Kolmogorov-Arnold Networks (ConvKANs), integrating learnable spline-based activation functions into convolutional layers, for PD classification using structural MRI. The first 3D implementation of ConvKANs for medical imaging is presented, comparing their performance to Convolutional Neural Networks (CNNs) and Graph Convolutional Networks (GCNs) across three open-source datasets. Isolated analyses assessed performance within individual datasets, using cross-validation techniques. Holdout analyses evaluated cross-dataset generalizability by training models on two datasets and testing on the third, mirroring real-world clinical scenarios. In isolated analyses, 2D ConvKANs achieved the highest AUC of 0.99 (95% CI: 0.98-0.99) on the PPMI dataset, outperforming 2D CNNs (AUC: 0.97, p = 0.0092). 3D models showed promise, with 3D CNN and 3D ConvKAN reaching an AUC of 0.85 on PPMI. In holdout analyses, 3D ConvKAN demonstrated superior generalization, achieving an AUC of 0.85 on early-stage PD data. GCNs underperformed in 2D but improved in 3D implementations. These findings highlight ConvKANs' potential for PD detection, emphasize the importance of 3D analysis in capturing subtle brain changes, and underscore cross-dataset generalization challenges. This study advances AI-assisted PD diagnosis using structural MRI and emphasizes the need for larger-scale validation., Comment: 7 figures

Social dynamics are often driven by both pairwise (i.e., dyadic) relationships and higher-order (i.e., polyadic) group relationships, which one can describe using hypergraphs. To gain insight into the impact of polyadic relationships on dynamical processes on networks, we formulate and study a polyadic voter process, which we call the group-driven voter model (GVM), in which we incorporate the effect of group interactions by nonlinear interactions that are subject to a group (i.e., hyperedge) constraint. By examining the competition between nonlinearity and group sizes, we show that the GVM achieves consensus faster than standard voter-model dynamics, with an optimum minimizing exit time {\tau} . We substantiate this finding by using mean-field theory on annealed uniform hypergraphs with N nodes, for which {\tau} scales as A ln N, where the prefactor A depends both on the nonlinearity and on group-constraint factors. Our results reveal how competition between group interactions and nonlinearity shapes GVM dynamics. We thereby highlight the importance of such competing effects in complex systems with polyadic interactions., Comment: 6 pages, 5 figures

The advent of Ethereum 2.0 has introduced significant changes, particularly the shift to Proof-of-Stake consensus. This change presents new opportunities and challenges for arbitrage. Amidst these changes, we introduce BriDe Arbitrager, a novel tool designed for Ethereum 2.0 that leverages Bribery-driven attacks to Delay block production and increase arbitrage gains. The main idea is to allow malicious proposers to delay block production by bribing validators/proposers, thereby gaining more time to identify arbitrage opportunities. Through analysing the bribery process, we design an adaptive bribery strategy. Additionally, we propose a Delayed Transaction Ordering Algorithm to leverage the delayed time to amplify arbitrage profits for malicious proposers. To ensure fairness and automate the bribery process, we design and implement a bribery smart contract and a bribery client. As a result, BriDe Arbitrager enables adversaries controlling a limited (< 1/4) fraction of the voting powers to delay block production via bribery and arbitrage more profit. Extensive experimental results based on Ethereum historical transactions demonstrate that BriDe Arbitrager yields an average of 8.66 ETH (16,442.23 USD) daily profits. Furthermore, our approach does not trigger any slashing mechanisms and remains effective even under Proposer Builder Separation and other potential mechanisms will be adopted by Ethereum.

Sharding enhances blockchain scalability by partitioning nodes into multiple groups for concurrent transaction processing. Configuring a large number of \emph{small shards} helps improve the transaction concurrency of a sharding system. However, it increases the fraction of malicious nodes within each shard, easily leading to shard corruption and jeopardizing system security. Some existing works have attempted to improve concurrency by reducing the shard size while maintaining security. However, they often require frequent and time-consuming recovery of corrupted shards, leading to severe system stagnation. Also, they usually require network-wide consensus to guarantee security, which limits scalability. To address these issues, we propose DL-Chain, a blockchain sharding system that can securely provide \emph{high concurrency with stable and scalable performance.} Our core idea is a \underline{D}ual-\underline{L}ayer architecture and consensus, which consists of numerous smaller proposer shards (PSs) for transaction processing and multiple larger finalizer committees (FCs) for transaction finalization. To avoid system stagnation and thus guarantee stable performance, we ensure PSs' liveness even if they are corrupted through the cooperation of PSs and FCs, thus eliminating the recovery process of corrupted PSs. To better trade-off security and scalability, we fine-tune the FCs to enable multiple FCs to coexist securely. As a result, DL-Chain allows a larger fraction of malicious nodes in each PS ($<1/2$) and thus can securely configure smaller shards for boosted stable and scalable concurrency. Evaluation results show that DL-Chain achieves up to 10 times improvement in throughput compared to existing solutions and provides stable concurrency with up to 2,550 nodes.

Simulating time evolution is one of the most natural applications of quantum computers and is thus one of the most promising prospects for achieving practical quantum advantage. Here we develop quantum algorithms to extract thermodynamic properties by estimating the density of states (DOS), a central object in quantum statistical mechanics. We introduce key innovations that significantly improve the practicality and extend the generality of previous techniques. First, our approach allows one to estimate the DOS for a specific subspace of the full Hilbert space. This is crucial for fermionic systems, since fermion-to-qubit mappings partition the full Hilbert space into subspaces of fixed number, on which both canonical and grand canonical ensemble properties depend. Second, in our approach, by time evolving very simple, random initial states (e.g. random computational basis states), we can exactly recover the DOS on average. Third, due to circuit-depth limitations, we only reconstruct the DOS up to a convolution with a Gaussian window - thus all imperfections that shift the energy levels by less than the width of the convolution window will not significantly affect the estimated DOS. For these reasons we find the approach is a promising candidate for early quantum advantage as even short-time, noisy dynamics yield a semi-quantitative reconstruction of the DOS (convolution with a broad Gaussian window), while early fault tolerant devices will likely enable higher resolution DOS reconstruction through longer time evolution. We demonstrate the practicality of our approach in representative Fermi-Hubbard and spin models and find that our approach is highly robust to algorithmic errors in the time evolution and to gate noise. We show that our approach is compatible with NISQ-friendly variational methods, introducing a new technique for variational time evolution in noisy DOS computations., Comment: 13 pages main text, 6 pages appendices