Although large language models (LLMs) often produce impressive outputs, it remains unclear how they perform in real-world scenarios requiring strong reasoning skills and expert domain knowledge. We set out to investigate whether close- and open-source models (GPT-3.5, LLama-2, etc.) can be applied to answer and reason about difficult real-world-based questions. We focus on three popular medical benchmarks (MedQA-USMLE, MedMCQA, and PubMedQA) and multiple prompting scenarios: Chain-of-Thought (CoT, think step-by-step), few-shot and retrieval augmentation. Based on an expert annotation of the generated CoTs, we found that InstructGPT can often read, reason and recall expert knowledge. Last, by leveraging advances in prompt engineering (few-shot and ensemble methods), we demonstrated that GPT-3.5 not only yields calibrated predictive distributions, but also reaches the passing score on three datasets: MedQA-USMLE 60.2%, MedMCQA 62.7% and PubMedQA 78.2%. Open-source models are closing the gap: Llama-2 70B also passed the MedQA-USMLE with 62.5% accuracy.
相關內容
大語言模型是基于海量文本數據訓練的深度學習模型。它不僅能夠生成自然語言文本，還能夠深入理解文本含義，處理各種自然語言任務，如文本摘要、問答、翻譯等。2023年，大語言模型及其在人工智能領域的應用已成為全球科技研究的熱點，其在規模上的增長尤為引人注目，參數量已從最初的十幾億躍升到如今的一萬億。參數量的提升使得模型能夠更加精細地捕捉人類語言微妙之處，更加深入地理解人類語言的復雜性。在過去的一年里，大語言模型在吸納新知識、分解復雜任務以及圖文對齊等多方面都有顯著提升。隨著技術的不斷成熟，它將不斷拓展其應用范圍，為人類提供更加智能化和個性化的服務，進一步改善人們的生活和生產方式。
Cellular networks have become one of the critical infrastructures, as many services depend increasingly on wireless connectivity. Therefore, it is important to quantify the resilience of existing cellular network infrastructures against potential risks, ranging from natural disasters to security attacks, that might occur with a low probability but can lead to severe disruption of the services. In this paper, we combine models with public data from national bodies on mobile network operator (MNO) infrastructures, population distribution, and urbanity level to assess the coverage and capacity of a cellular network at a country scale. Our analysis offers insights on the potential weak points that need improvement to ensure a low fraction of disconnected population (FDP) and high fraction of satisfied population (FSP). As a resilience improvement approach, we investigate in which regions and to what extent each MNO can benefit from infrastructure sharing or national roaming, i.e., all MNOs act as a single national operator. As our case study, we focus on Dutch cellular infrastructure and model risks as random failures, correlated failures in a geographic region, and abrupt increase in the number of users. Our analysis shows that there is a wide performance difference across MNOs and geographic regions in terms of FDP and FSP. However, national roaming consistently offers significant benefits, e.g., up to 13% improvement in FDP and up to 55% in FSP when the networks function without any failures.
Experimentation is an intrinsic part of research in artificial intelligence since it allows for collecting quantitative observations, validating hypotheses, and providing evidence for their reformulation. For that reason, experimentation must be coherent with the purposes of the research, properly addressing the relevant questions in each case. Unfortunately, the literature is full of works whose experimentation is neither rigorous nor convincing, oftentimes designed to support prior beliefs rather than answering the relevant research questions. In this paper, we focus on the field of metaheuristic optimization, since it is our main field of work, and it is where we have observed the misconduct that has motivated this letter. Even if we limit the focus of this manuscript to the experimental part of the research, our main goal is to sew the seed of sincere critical assessment of our work, sparking a reflection process both at the individual and the community level. Such a reflection process is too complex and extensive to be tackled as a whole. Therefore, to bring our feet to the ground, we will include in this document our reflections about the role of experimentation in our work, discussing topics such as the use of benchmark instances vs instance generators, or the statistical assessment of empirical results. That is, all the statements included in this document are personal views and opinions, which can be shared by others or not. Certainly, having different points of view is the basis to establish a good discussion process.
The integration of large language models (LLMs) into the medical field has gained significant attention due to their promising accuracy in simulated clinical decision-making settings. However, clinical decision-making is more complex than simulations because physicians' decisions are shaped by many factors, including the presence of cognitive bias. However, the degree to which LLMs are susceptible to the same cognitive biases that affect human clinicians remains unexplored. Our hypothesis posits that when LLMs are confronted with clinical questions containing cognitive biases, they will yield significantly less accurate responses compared to the same questions presented without such biases.In this study, we developed BiasMedQA, a novel benchmark for evaluating cognitive biases in LLMs applied to medical tasks. Using BiasMedQA we evaluated six LLMs, namely GPT-4, Mixtral-8x70B, GPT-3.5, PaLM-2, Llama 2 70B-chat, and the medically specialized PMC Llama 13B. We tested these models on 1,273 questions from the US Medical Licensing Exam (USMLE) Steps 1, 2, and 3, modified to replicate common clinically-relevant cognitive biases. Our analysis revealed varying effects for biases on these LLMs, with GPT-4 standing out for its resilience to bias, in contrast to Llama 2 70B-chat and PMC Llama 13B, which were disproportionately affected by cognitive bias. Our findings highlight the critical need for bias mitigation in the development of medical LLMs, pointing towards safer and more reliable applications in healthcare.
Large language models (LLMs) have been extensively studied for their abilities to generate convincing natural language sequences, however their utility for quantitative information retrieval is less well understood. In this paper we explore the feasibility of LLMs as a mechanism for quantitative knowledge retrieval to aid data analysis tasks such as elicitation of prior distributions for Bayesian models and imputation of missing data. We present a prompt engineering framework, treating an LLM as an interface to a latent space of scientific literature, comparing responses in different contexts and domains against more established approaches. Implications and challenges of using LLMs as 'experts' are discussed.
Although large language models (LLMs) are reshaping various aspects of human life, our current understanding of their impacts remains somewhat constrained. Here we investigate the impact of LLMs on human communication, using data on consumer complaints in the financial industry. By employing an AI detection tool on more than 820K complaints gathered by the Consumer Financial Protection Bureau (CFPB), we find a sharp increase in the likely use of LLMs shortly after the release of ChatGPT. Moreover, the likely LLM usage was positively correlated with message persuasiveness (i.e., increased likelihood of obtaining relief from financial firms). Computational linguistic analyses suggest that the positive correlation may be explained by LLMs' enhancement of various linguistic features. Based on the results of these observational studies, we hypothesize that LLM usage may enhance a comprehensive set of linguistic features, increasing message persuasiveness to receivers with heterogeneous linguistic preferences (i.e., linguistic feature alignment). We test this hypothesis in preregistered experiments and find support for it. As an instance of early empirical demonstrations of LLM usage for enhancing persuasion, our research highlights the transformative potential of LLMs in human communication.
The evaluation of text-generative vision-language models is a challenging yet crucial endeavor. By addressing the limitations of existing Visual Question Answering (VQA) benchmarks and proposing innovative evaluation methodologies, our research seeks to advance our understanding of these models' capabilities. We propose a novel VQA benchmark based on well-known visual classification datasets which allows a granular evaluation of text-generative vision-language models and their comparison with discriminative vision-language models. To improve the assessment of coarse answers on fine-grained classification tasks, we suggest using the semantic hierarchy of the label space to ask automatically generated follow-up questions about the ground-truth category. Finally, we compare traditional NLP and LLM-based metrics for the problem of evaluating model predictions given ground-truth answers. We perform a human evaluation study upon which we base our decision on the final metric. We apply our benchmark to a suite of vision-language models and show a detailed comparison of their abilities on object, action, and attribute classification. Our contributions aim to lay the foundation for more precise and meaningful assessments, facilitating targeted progress in the exciting field of vision-language modeling.
The stochastic simulation algorithm (SSA) and the corresponding Monte Carlo (MC) method are among the most common approaches for studying stochastic processes. They rely on knowledge of interevent probability density functions (PDFs) and on information about dependencies between all possible events. Analytical representations of a PDF are difficult to specify in advance, in many real life applications. Knowing the shapes of PDFs, and using experimental data, different optimization schemes can be applied in order to evaluate probability density functions and, therefore, the properties of the studied system. Such methods, however, are computationally demanding, and often not feasible. We show that, in the case where experimentally accessed properties are directly related to the frequencies of events involved, it may be possible to replace the heavy Monte Carlo core of optimization schemes with an analytical solution. Such a replacement not only provides a more accurate estimation of the properties of the process, but also reduces the simulation time by a factor of order of the sample size (at least $\approx 10^4$). The proposed analytical approach is valid for any choice of PDF. The accuracy, computational efficiency, and advantages of the method over MC procedures are demonstrated in the exactly solvable case and in the evaluation of branching fractions in controlled radical polymerization (CRP) of acrylic monomers. This polymerization can be modeled by a constrained stochastic process. Constrained systems are quite common, and this makes the method useful for various applications.
Incorporating prior knowledge into pre-trained language models has proven to be effective for knowledge-driven NLP tasks, such as entity typing and relation extraction. Current pre-training procedures usually inject external knowledge into models by using knowledge masking, knowledge fusion and knowledge replacement. However, factual information contained in the input sentences have not been fully mined, and the external knowledge for injecting have not been strictly checked. As a result, the context information cannot be fully exploited and extra noise will be introduced or the amount of knowledge injected is limited. To address these issues, we propose MLRIP, which modifies the knowledge masking strategies proposed by ERNIE-Baidu, and introduce a two-stage entity replacement strategy. Extensive experiments with comprehensive analyses illustrate the superiority of MLRIP over BERT-based models in military knowledge-driven NLP tasks.
This paper does not describe a working system. Instead, it presents a single idea about representation which allows advances made by several different groups to be combined into an imaginary system called GLOM. The advances include transformers, neural fields, contrastive representation learning, distillation and capsules. GLOM answers the question: How can a neural network with a fixed architecture parse an image into a part-whole hierarchy which has a different structure for each image? The idea is simply to use islands of identical vectors to represent the nodes in the parse tree. If GLOM can be made to work, it should significantly improve the interpretability of the representations produced by transformer-like systems when applied to vision or language
Recent advances in 3D fully convolutional networks (FCN) have made it feasible to produce dense voxel-wise predictions of volumetric images. In this work, we show that a multi-class 3D FCN trained on manually labeled CT scans of several anatomical structures (ranging from the large organs to thin vessels) can achieve competitive segmentation results, while avoiding the need for handcrafting features or training class-specific models. To this end, we propose a two-stage, coarse-to-fine approach that will first use a 3D FCN to roughly define a candidate region, which will then be used as input to a second 3D FCN. This reduces the number of voxels the second FCN has to classify to ~10% and allows it to focus on more detailed segmentation of the organs and vessels. We utilize training and validation sets consisting of 331 clinical CT images and test our models on a completely unseen data collection acquired at a different hospital that includes 150 CT scans, targeting three anatomical organs (liver, spleen, and pancreas). In challenging organs such as the pancreas, our cascaded approach improves the mean Dice score from 68.5 to 82.2%, achieving the highest reported average score on this dataset. We compare with a 2D FCN method on a separate dataset of 240 CT scans with 18 classes and achieve a significantly higher performance in small organs and vessels. Furthermore, we explore fine-tuning our models to different datasets. Our experiments illustrate the promise and robustness of current 3D FCN based semantic segmentation of medical images, achieving state-of-the-art results. Our code and trained models are available for download: //github.com/holgerroth/3Dunet_abdomen_cascade.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191