Search

Large language models (LLMs) have been widely adopted in various downstream task domains. However, their ability to directly recall and apply factual medical knowledge remains under-explored. Most existing medical QA benchmarks assess complex reasoning or multi-hop inference, making it difficult to isolate LLMs' inherent medical knowledge from their reasoning capabilities. Given the high-stakes nature of medical applications, where incorrect information can have critical consequences, it is essential to evaluate how well LLMs encode, retain, and recall fundamental medical facts. To bridge this gap, we introduce the Medical Knowledge Judgment, a dataset specifically designed to measure LLMs' one-hop factual medical knowledge. MKJ is constructed from the Unified Medical Language System (UMLS), a large-scale repository of standardized biomedical vocabularies and knowledge graphs. We frame knowledge assessment as a binary judgment task, requiring LLMs to verify the correctness of medical statements extracted from reliable and structured knowledge sources. Our experiments reveal that LLMs struggle with factual medical knowledge retention, exhibiting significant performance variance across different semantic categories, particularly for rare medical conditions. Furthermore, LLMs show poor calibration, often being overconfident in incorrect answers. To mitigate these issues, we explore retrieval-augmented generation, demonstrating its effectiveness in improving factual accuracy and reducing uncertainty in medical decision-making., Comment: 15 pages, 11 figures

Large language models (LLMs) excel across various tasks, but standard first-order (FO) fine-tuning demands considerable memory, significantly limiting real-world deployment. Recently, zeroth-order (ZO) optimization stood out as a promising memory-efficient training paradigm, avoiding backward passes and relying solely on forward passes for gradient estimation, making it attractive for resource-constrained scenarios. However, ZO method lags far behind FO method in both convergence speed and accuracy. To bridge the gap, we introduce a novel layer-wise divergence analysis that uncovers the distinct update pattern of FO and ZO optimization. Aiming to resemble the learning capacity of FO method from the findings, we propose \textbf{Di}vergence-driven \textbf{Z}eroth-\textbf{O}rder (\textbf{DiZO}) optimization. DiZO conducts divergence-driven layer adaptation by incorporating projections to ZO updates, generating diverse-magnitude updates precisely scaled to layer-wise individual optimization needs. Our results demonstrate that DiZO significantly reduces the needed iterations for convergence without sacrificing throughput, cutting training GPU hours by up to 48\% on various datasets. Moreover, DiZO consistently outperforms the representative ZO baselines in fine-tuning RoBERTa-large, OPT-series, and Llama-series on downstream tasks and, in some cases, even surpasses memory-intensive FO fine-tuning.

Constrained optimization demands highly efficient solvers which promotes the development of learn-to-optimize (L2O) approaches. As a data-driven method, L2O leverages neural networks to efficiently produce approximate solutions. However, a significant challenge remains in ensuring both optimality and feasibility of neural networks' output. To tackle this issue, we introduce Homeomorphic Polar Learning (HoP) to solve the star-convex hard-constrained optimization by embedding homeomorphic mapping in neural networks. The bijective structure enables end-to-end training without extra penalty or correction. For performance evaluation, we evaluate HoP's performance across a variety of synthetic optimization tasks and real-world applications in wireless communications. In all cases, HoP achieves solutions closer to the optimum than existing L2O methods while strictly maintaining feasibility., Comment: in submission

Large language models (LLMs) exhibit remarkable capabilities in visual inspection of medical time-series data, achieving proficiency comparable to human clinicians. However, their broad scope limits domain-specific precision, and proprietary weights hinder fine-tuning for specialized datasets. In contrast, small specialized models (SSMs) excel in targeted tasks but lack the contextual reasoning required for complex clinical decision-making. To address these challenges, we propose ConMIL (Conformalized Multiple Instance Learning), a decision-support SSM that integrates seamlessly with LLMs. By using Multiple Instance Learning (MIL) to identify clinically significant signal segments and conformal prediction for calibrated set-valued outputs, ConMIL enhances LLMs' interpretative capabilities for medical time-series analysis. Experimental results demonstrate that ConMIL significantly improves the performance of state-of-the-art LLMs, such as ChatGPT4.0 and Qwen2-VL-7B. Specifically, \ConMIL{}-supported Qwen2-VL-7B achieves 94.92% and 96.82% precision for confident samples in arrhythmia detection and sleep staging, compared to standalone LLM accuracy of 46.13% and 13.16%. These findings highlight the potential of ConMIL to bridge task-specific precision and broader contextual reasoning, enabling more reliable and interpretable AI-driven clinical decision support.

With the rapid advancements in large language model (LLM) technology and the emergence of bioinformatics-specific language models (BioLMs), there is a growing need for a comprehensive analysis of the current landscape, computational characteristics, and diverse applications. This survey aims to address this need by providing a thorough review of BioLMs, focusing on their evolution, classification, and distinguishing features, alongside a detailed examination of training methodologies, datasets, and evaluation frameworks. We explore the wide-ranging applications of BioLMs in critical areas such as disease diagnosis, drug discovery, and vaccine development, highlighting their impact and transformative potential in bioinformatics. We identify key challenges and limitations inherent in BioLMs, including data privacy and security concerns, interpretability issues, biases in training data and model outputs, and domain adaptation complexities. Finally, we highlight emerging trends and future directions, offering valuable insights to guide researchers and clinicians toward advancing BioLMs for increasingly sophisticated biological and clinical applications., Comment: 64 pages, 1 figure

Artificial Intelligence (AI) has become essential in modern healthcare, with large language models (LLMs) offering promising advances in clinical decision-making. Traditional model-based approaches, including those leveraging in-context demonstrations and those with specialized medical fine-tuning, have demonstrated strong performance in medical language processing but struggle with real-time adaptability, multi-step reasoning, and handling complex medical tasks. Agent-based AI systems address these limitations by incorporating reasoning traces, tool selection based on context, knowledge retrieval, and both short- and long-term memory. These additional features enable the medical AI agent to handle complex medical scenarios where decision-making should be built on real-time interaction with the environment. Therefore, unlike conventional model-based approaches that treat medical queries as isolated questions, medical AI agents approach them as complex tasks and behave more like human doctors. In this paper, we study the choice of the backbone LLM for medical AI agents, which is the foundation for the agent's overall reasoning and action generation. In particular, we consider the emergent o1 model and examine its impact on agents' reasoning, tool-use adaptability, and real-time information retrieval across diverse clinical scenarios, including high-stakes settings such as intensive care units (ICUs). Our findings demonstrate o1's ability to enhance diagnostic accuracy and consistency, paving the way for smarter, more responsive AI tools that support better patient outcomes and decision-making efficacy in clinical practice.

Fine-tuning large language models (LLMs) poses significant memory challenges, as the back-propagation process demands extensive resources, especially with growing model sizes. Recent work, MeZO, addresses this issue using a zeroth-order (ZO) optimization method, which reduces memory consumption by matching the usage to the inference phase. However, MeZO experiences slow convergence due to varying curvatures across model parameters. To overcome this limitation, we introduce HELENE, a novel scalable and memory-efficient optimizer that integrates annealed A-GNB gradients with a diagonal Hessian estimation and layer-wise clipping, serving as a second-order pre-conditioner. This combination allows for faster and more stable convergence. Our theoretical analysis demonstrates that HELENE improves convergence rates, particularly for models with heterogeneous layer dimensions, by reducing the dependency on the total parameter space dimension. Instead, the method scales with the largest layer dimension, making it highly suitable for modern LLM architectures. Experimental results on RoBERTa-large and OPT-1.3B across multiple tasks show that HELENE achieves up to a 20x speedup compared to MeZO, with average accuracy improvements of 1.5%. Furthermore, HELENE remains compatible with both full parameter tuning and parameter-efficient fine-tuning (PEFT), outperforming several state-of-the-art optimizers. The codes will be released after reviewing.

As artificial intelligence (AI) continues to advance, it demonstrates capabilities comparable to human intelligence, with significant potential to transform education and workforce development. This study evaluates OpenAI o1-preview's ability to perform higher-order cognitive tasks across 14 dimensions, including critical thinking, systems thinking, computational thinking, design thinking, metacognition, data literacy, creative thinking, abstract reasoning, quantitative reasoning, logical reasoning, analogical reasoning, and scientific reasoning. We used validated instruments like the Ennis-Weir Critical Thinking Essay Test and the Biological Systems Thinking Test to compare the o1-preview's performance with human performance systematically. Our findings reveal that o1-preview outperforms humans in most categories, achieving 150% better results in systems thinking, computational thinking, data literacy, creative thinking, scientific reasoning, and abstract reasoning. However, compared to humans, it underperforms by around 25% in logical reasoning, critical thinking, and quantitative reasoning. In analogical reasoning, both o1-preview and humans achieved perfect scores. Despite these strengths, the o1-preview shows limitations in abstract reasoning, where human psychology students outperform it, highlighting the continued importance of human oversight in tasks requiring high-level abstraction. These results have significant educational implications, suggesting a shift toward developing human skills that complement AI, such as creativity, abstract reasoning, and critical thinking. This study emphasizes the transformative potential of AI in education and calls for a recalibration of educational goals, teaching methods, and curricula to align with an AI-driven world., Comment: An assessment of OpenAI o1-Preview for Higher Order Thinking in Education

This comprehensive study evaluates the performance of OpenAI's o1-preview large language model across a diverse array of complex reasoning tasks, spanning multiple domains, including computer science, mathematics, natural sciences, medicine, linguistics, and social sciences. Through rigorous testing, o1-preview demonstrated remarkable capabilities, often achieving human-level or superior performance in areas ranging from coding challenges to scientific reasoning and from language processing to creative problem-solving. Key findings include: -83.3% success rate in solving complex competitive programming problems, surpassing many human experts. -Superior ability in generating coherent and accurate radiology reports, outperforming other evaluated models. -100% accuracy in high school-level mathematical reasoning tasks, providing detailed step-by-step solutions. -Advanced natural language inference capabilities across general and specialized domains like medicine. -Impressive performance in chip design tasks, outperforming specialized models in areas such as EDA script generation and bug analysis. -Remarkable proficiency in anthropology and geology, demonstrating deep understanding and reasoning in these specialized fields. -Strong capabilities in quantitative investing. O1 has comprehensive financial knowledge and statistical modeling skills. -Effective performance in social media analysis, including sentiment analysis and emotion recognition. The model excelled particularly in tasks requiring intricate reasoning and knowledge integration across various fields. While some limitations were observed, including occasional errors on simpler problems and challenges with certain highly specialized concepts, the overall results indicate significant progress towards artificial general intelligence.

Face recognition pipelines have been widely deployed in various mission-critical systems in trust, equitable and responsible AI applications. However, the emergence of adversarial attacks has threatened the security of the entire recognition pipeline. Despite the sheer number of attack methods proposed for crafting adversarial examples in both digital and physical forms, it is never an easy task to assess the real threat level of different attacks and obtain useful insight into the key risks confronted by face recognition systems. Traditional attacks view imperceptibility as the most important measurement to keep perturbations stealthy, while we suspect that industry professionals may possess a different opinion. In this paper, we delve into measuring the threat brought about by adversarial attacks from the perspectives of the industry and the applications of face recognition. In contrast to widely studied sophisticated attacks in the field, we propose an effective yet easy-to-launch physical adversarial attack, named AdvColor, against black-box face recognition pipelines in the physical world. AdvColor fools models in the recognition pipeline via directly supplying printed photos of human faces to the system under adversarial illuminations. Experimental results show that physical AdvColor examples can achieve a fooling rate of more than 96% against the anti-spoofing model and an overall attack success rate of 88% against the face recognition pipeline. We also conduct a survey on the threats of prevailing adversarial attacks, including AdvColor, to understand the gap between the machine-measured and human-assessed threat levels of different forms of adversarial attacks. The survey results surprisingly indicate that, compared to deliberately launched imperceptible attacks, perceptible but accessible attacks pose more lethal threats to real-world commercial systems of face recognition., Comment: 19 pages, 12 figures

The increasing interest in hydrogen as a clean energy source has led to extensive research into its transmission, storage, and integration with bulk power systems. With the evolution of hydrogen technologies towards greater efficiency, and cost-effectiveness, it becomes essential to examine the operation and expansion of grids that include both electric power and hydrogen facilities. This paper introduces an expansion strategy for electric power and hydrogen transmission systems, tailored for future renewable energy-enriched grids. Our proposed transmission expansion planning with hydrogen facilities (TEP-H) model integrates daily operations of both electric power and hydrogen transmissions. The fuel cells and electrolyzers are used for electrical-hydrogen energy conversion, and related constraints are considered in TEP-H. We applied TEP-H to the Texas 123-bus backbone transmission grid (TX-123BT), for various renewable penetration levels and hydrogen technology development assumptions. It gave us insights on the scenarios that hydrogen transmission become feasible and economically beneficial. We also compared the performance of TX-123BT system with the hybrid transmission investment and the pure electrical transmission investment obtained by a traditional transmission expansion planning (TEP-T) model. The numerical results indicate that future renewable grids can have lower total cost with TEP-H if future electrical-hydrogen energy conversion efficiency is high., Comment: 9 pages

Video recognition systems are vulnerable to adversarial examples. Recent studies show that style transfer-based and patch-based unrestricted perturbations can effectively improve attack efficiency. These attacks, however, face two main challenges: 1) Adding large stylized perturbations to all pixels reduces the naturalness of the video and such perturbations can be easily detected. 2) Patch-based video attacks are not extensible to targeted attacks due to the limited search space of reinforcement learning that has been widely used in video attacks recently. In this paper, we focus on the video black-box setting and propose a novel attack framework named LogoStyleFool by adding a stylized logo to the clean video. We separate the attack into three stages: style reference selection, reinforcement-learning-based logo style transfer, and perturbation optimization. We solve the first challenge by scaling down the perturbation range to a regional logo, while the second challenge is addressed by complementing an optimization stage after reinforcement learning. Experimental results substantiate the overall superiority of LogoStyleFool over three state-of-the-art patch-based attacks in terms of attack performance and semantic preservation. Meanwhile, LogoStyleFool still maintains its performance against two existing patch-based defense methods. We believe that our research is beneficial in increasing the attention of the security community to such subregional style transfer attacks., Comment: 14 pages, 3 figures. Accepted to AAAI 2024

As renewable energy is becoming the major resource in future grids, the weather and climate can have a higher impact on grid reliability. Transmission expansion planning (TEP) has the potential to reinforce a transmission network that is suitable for climate-impacted grids. In this paper, we propose a systematic TEP procedure for climate-impacted renewable energy-enriched grids. Particularly, this work developed an improved model for TEP considering climate impact (TEP-CI) and evaluated the system reliability with the obtained transmission investment plan. Firstly, we created climate-impacted spatio-temporal future grid data to facilitate the TEP-CI study, which includes the future climate-dependent renewable production as well as the dynamic rating profiles of the Texas 123-bus backbone transmission system (TX-123BT). Secondly, we proposed the TEP-CI which considers the variation in renewable production and dynamic line rating, and obtained the investment plan for future TX-123BT. Thirdly, we presented a customized security-constrained unit commitment (SCUC) specifically for climate-impacted grids. The future grid reliability under various investment scenarios is analyzed based on the daily operation conditions from SCUC simulations. The whole procedure presented in this paper enables numerical studies on grid planning considering climate impacts. It can also serve as a benchmark for other TEP-CI research and performance evaluation., Comment: 11 pages, 8 figures

Accurate load forecasting is critical for efficient and reliable operations of the electric power system. A large part of electricity consumption is affected by weather conditions, making weather information an important determinant of electricity usage. Personal appliances and industry equipment also contribute significantly to electricity demand with temporal patterns, making time a useful factor to consider in load forecasting. This work develops several machine learning (ML) models that take various time and weather information as part of the input features to predict the short-term system-wide total load. Ablation studies were also performed to investigate and compare the impacts of different weather factors on the prediction accuracy. Actual load and historical weather data for the same region were processed and then used to train the ML models. It is interesting to observe that using all available features, each of which may be correlated to the load, is unlikely to achieve the best forecasting performance; features with redundancy may even decrease the inference capabilities of ML models. This indicates the importance of feature selection for ML models. Overall, case studies demonstrated the effectiveness of ML models trained with different weather and time input features for ERCOT load forecasting.

Detecting weak target is an important and challenging problem in many applications such as radar, sonar etc. However, conventional detection methods are often ineffective in this case because of low signal-to-noise ratio (SNR). This paper presents a track-before-detect (TBD) algorithm based on an improved particle filter, i.e. cost-reference particle filter bank (CRPFB), which turns the problem of target detection to the problem of two-layer hypothesis testing. The first layer is implemented by CRPFB for state estimation of possible target. CRPFB has entirely parallel structure, consisting amounts of cost-reference particle filters with different hypothesized prior information. The second layer is to compare a test metric with a given threshold, which is constructed from the output of the first layer and fits GEV distribution. The performance of our proposed TBD algorithm and the existed TBD algorithms are compared according to the experiments on nonlinear frequency modulated (NLFM) signal detection and tracking. Simulation results show that the proposed TBD algorithm has better performance than the state-of-the-arts in detection, tracking, and time efficiency.

Artificial General Intelligence (AGI), possessing the capacity to comprehend, learn, and execute tasks with human cognitive abilities, engenders significant anticipation and intrigue across scientific, commercial, and societal arenas. This fascination extends particularly to the Internet of Things (IoT), a landscape characterized by the interconnection of countless devices, sensors, and systems, collectively gathering and sharing data to enable intelligent decision-making and automation. This research embarks on an exploration of the opportunities and challenges towards achieving AGI in the context of the IoT. Specifically, it starts by outlining the fundamental principles of IoT and the critical role of Artificial Intelligence (AI) in IoT systems. Subsequently, it delves into AGI fundamentals, culminating in the formulation of a conceptual framework for AGI's seamless integration within IoT. The application spectrum for AGI-infused IoT is broad, encompassing domains ranging from smart grids, residential environments, manufacturing, and transportation to environmental monitoring, agriculture, healthcare, and education. However, adapting AGI to resource-constrained IoT settings necessitates dedicated research efforts. Furthermore, the paper addresses constraints imposed by limited computing resources, intricacies associated with large-scale IoT communication, as well as the critical concerns pertaining to security and privacy.

On-chip microring resonators (MRRs) have been proposed to construct the time-delayed reservoir computing (RC), which offers promising configurations available for computation with high scalability, high-density computing, and easy fabrication. A single MRR, however, is inadequate to supply enough memory for the computational task with diverse memory requirements. Large memory needs are met by the MRR with optical feedback waveguide, but at the expense of its large footprint. In the structure, the ultra-long optical feedback waveguide substantially limits the scalable photonic RC integrated designs. In this paper, a time-delayed RC is proposed by utilizing a silicon-based nonlinear MRR in conjunction with an array of linear MRRs. These linear MRRs possess a high quality factor, providing sufficient memory capacity for the entire system. We quantitatively analyze and assess the proposed RC structure's performance on three classical tasks with diverse memory requirements, i.e., the Narma 10, Mackey-Glass, and Santa Fe chaotic timeseries prediction tasks. The proposed system exhibits comparable performance to the MRR with an ultra-long optical feedback waveguide-based system when it comes to handling the Narma 10 task, which requires a significant memory capacity. Nevertheless, the overall length of these linear MRRs is significantly smaller, by three orders of magnitude, compared to the ultra-long feedback waveguide in the MRR with optical feedback waveguide-based system. The compactness of this structure has significant implications for the scalability and seamless integration of photonic RC.