With the increasing size of pre-trained language models (PLMs), fine-tuning all the parameters in the model is not efficient, especially when there are a large number of downstream tasks, which incur significant training and storage costs. Many parameter-efficient fine-tuning (PEFT) approaches have been proposed, among which, Low-Rank Adaptation (LoRA) is a representative approach that injects trainable rank decomposition matrices into every target module. Yet LoRA ignores the importance of parameters in different modules. To address this problem, many works have been proposed to prune the parameters of LoRA. However, under limited training conditions, the upper bound of the rank of the pruned parameter matrix is still affected by the preset values. We, therefore, propose IncreLoRA, an incremental parameter allocation method that adaptively adds trainable parameters during training based on the importance scores of each module. This approach is different from the pruning method as it is not limited by the initial number of training parameters, and each parameter matrix has a higher rank upper bound for the same training overhead. We conduct extensive experiments on GLUE to demonstrate the effectiveness of IncreLoRA. The results show that our method owns higher parameter efficiency, especially when under the low-resource settings where our method significantly outperforms the baselines. Our code is publicly available.

Prostate cancer pathology plays a crucial role in clinical management but is time-consuming. Artificial intelligence (AI) shows promise in detecting prostate cancer and grading patterns. We tested an AI-based digital twin of a pathologist, vPatho, on 2,603 histology images of prostate tissue stained with hematoxylin and eosin. We analyzed various factors influencing tumor-grade disagreement between vPatho and six human pathologists. Our results demonstrated that vPatho achieved comparable performance in prostate cancer detection and tumor volume estimation, as reported in the literature. Concordance levels between vPatho and human pathologists were examined. Notably, moderate to substantial agreement was observed in identifying complementary histological features such as ductal, cribriform, nerve, blood vessels, and lymph cell infiltrations. However, concordance in tumor grading showed a decline when applied to prostatectomy specimens (kappa = 0.44) compared to biopsy cores (kappa = 0.70). Adjusting the decision threshold for the secondary Gleason pattern from 5% to 10% improved the concordance level between pathologists and vPatho for tumor grading on prostatectomy specimens (kappa from 0.44 to 0.64). Potential causes of grade discordance included the vertical extent of tumors toward the prostate boundary and the proportions of slides with prostate cancer. Gleason pattern 4 was particularly associated with discordance. Notably, grade discordance with vPatho was not specific to any of the six pathologists involved in routine clinical grading. In conclusion, our study highlights the potential utility of AI in developing a digital twin of a pathologist. This approach can help uncover limitations in AI adoption and the current grading system for prostate cancer pathology.

Studies on semi-supervised medical image segmentation (SSMIS) have seen fast progress recently. Due to the limited labelled data, SSMIS methods mainly focus on effectively leveraging unlabeled data to enhance the segmentation performance. However, despite their promising performance, current state-of-the-art methods often prioritize integrating complex techniques and loss terms rather than addressing the core challenges of semi-supervised scenarios directly. We argue that the key to SSMIS lies in generating substantial and appropriate prediction disagreement on unlabeled data. To this end, we emphasize the crutiality of data perturbation and model stabilization in semi-supervised segmentation, and propose a simple yet effective approach to boost SSMIS performance significantly, dubbed DPMS. Specifically, we first revisit SSMIS from three distinct perspectives: the data, the model, and the loss, and conduct a comprehensive study of corresponding strategies to examine their effectiveness. Based on these examinations, we then propose DPMS, which adopts a plain teacher-student framework with a standard supervised loss and unsupervised consistency loss. To produce appropriate prediction disagreements, DPMS perturbs the unlabeled data via strong augmentations to enlarge prediction disagreements considerably. On the other hand, using EMA teacher when strong augmentation is applied does not necessarily improve performance. DPMS further utilizes a forwarding-twice and momentum updating strategies for normalization statistics to stabilize the training on unlabeled data effectively. Despite its simplicity, DPMS can obtain new state-of-the-art performance on the public 2D ACDC and 3D LA datasets across various semi-supervised settings, e.g. obtaining a remarkable 22.62% improvement against previous SOTA on ACDC with 5% labels.

Masked autoencoder (MAE) is a promising self-supervised pre-training technique that can improve the representation learning of a neural network without human intervention. However, applying MAE directly to volumetric medical images poses two challenges: (i) a lack of global information that is crucial for understanding the clinical context of the holistic data, (ii) no guarantee of stabilizing the representations learned from randomly masked inputs. To address these limitations, we propose the \textbf{G}lobal-\textbf{L}ocal \textbf{M}asked \textbf{A}uto\textbf{E}ncoder (GL-MAE), a simple yet effective self-supervised pre-training strategy. In addition to reconstructing masked local views, as in previous methods, GL-MAE incorporates global context learning by reconstructing masked global views. Furthermore, a complete global view is integrated as an anchor to guide the reconstruction and stabilize the learning process through global-to-global consistency learning and global-to-local consistency learning. Finetuning results on multiple datasets demonstrate the superiority of our method over other state-of-the-art self-supervised algorithms, highlighting its effectiveness on versatile volumetric medical image segmentation tasks, even when annotations are scarce. Our codes and models will be released upon acceptance.

In this survey, we review the literature investigating participatory budgeting as a social choice problem. Participatory Budgeting (PB) is a democratic process in which citizens are asked to vote on how to allocate a given amount of public money to a set of projects. From a social choice perspective, it corresponds then to the problem of aggregating opinions about which projects should be funded, into a budget allocation satisfying a budget constraint. This problem has received substantial attention in recent years and the literature is growing at a fast pace. In this survey, we present the most important research directions from the literature, each time presenting a large set of representative results. We only focus on the indivisible case, that is, PB problems in which projects can either be fully funded or not at all. The aim of the survey is to present a comprehensive overview of the state of the research on PB. We aim at providing both a general overview of the main research questions that are being investigated, and formal and unified definitions of the most important technical concepts from the literature.

We explore the data-parallel acceleration of physics-informed machine learning (PIML) schemes, with a focus on physics-informed neural networks (PINNs) for multiple graphics processing units (GPUs) architectures. In order to develop scale-robust and high-throughput PIML models for sophisticated applications which may require a large number of training points (e.g., involving complex and high-dimensional domains, non-linear operators or multi-physics), we detail a novel protocol based on $h$-analysis and data-parallel acceleration through the Horovod training framework. The protocol is backed by new convergence bounds for the generalization error and the train-test gap. We show that the acceleration is straightforward to implement, does not compromise training, and proves to be highly efficient and controllable, paving the way towards generic scale-robust PIML. Extensive numerical experiments with increasing complexity illustrate its robustness and consistency, offering a wide range of possibilities for real-world simulations.

This paper examines the effectiveness of several forecasting methods for predicting inflation, focusing on aggregating disaggregated forecasts - also known in the literature as the bottom-up approach. Taking the Brazilian case as an application, we consider different disaggregation levels for inflation and employ a range of traditional time series techniques as well as linear and nonlinear machine learning (ML) models to deal with a larger number of predictors. For many forecast horizons, the aggregation of disaggregated forecasts performs just as well survey-based expectations and models that generate forecasts using the aggregate directly. Overall, ML methods outperform traditional time series models in predictive accuracy, with outstanding performance in forecasting disaggregates. Our results reinforce the benefits of using models in a data-rich environment for inflation forecasting, including aggregating disaggregated forecasts from ML techniques, mainly during volatile periods. Starting from the COVID-19 pandemic, the random forest model based on both aggregate and disaggregated inflation achieves remarkable predictive performance at intermediate and longer horizons.

Asynchronous online discussions are a common fundamental tools to facilitate social interaction in hybrid and online courses. However, instructors lack the tools to accomplish the overwhelming task of evaluating asynchronous online discussion activities. In this paper we present an approach that uses Latent Dirichlet Analysis (LDA) and the instructor's keywords to automatically extract codes from a relatively small dataset. We use the generated codes to build an Epistemic Network Analysis (ENA) model and compare this model with a previous ENA model built by human coders. The results show that there is no statistical difference between the two models. We present an analysis of these models and discuss the potential use of ENA as a visualization to help instructors evaluating asynchronous online discussions.

Wikidata is one of the largest public general-purpose Knowledge Bases (KBs). Yet, due to its collaborative nature, its schema and taxonomy have become convoluted. For the YAGO 4 KB, we combined Wikidata with the ontology from Schema.org, which reduced and cleaned up the taxonomy and constraints and made it possible to run automated reasoners on the data. However, it also cut away large parts of the Wikidata taxonomy. In this paper, we present our effort to merge the entire Wikidata taxonomy into the YAGO KB as much as possible. We pay particular attention to logical constraints and a careful distinction of classes and instances. Our work creates YAGO 4.5, which adds a rich layer of informative classes to YAGO, while at the same time keeping the KB logically consistent.

The strategy of training the model from scratch in a specific language or domain serves two essential purposes: i) enhancing performance in the particular linguistic or domain context, and ii) ensuring effective tokenization. The main limitation inherent to this approach lies in the associated cost, which can reach six to seven-digit dollar values, depending on the model size and the number of parameters involved. The main solution to overcome the cost challenge is to rely on available pre-trained models, which, despite recent advancements such as the LLaMA and LLaMA-2 models, still demonstrate inefficiency for certain specific domain problems or prove ineffective in scenarios involving conversational memory resources, given the large number of tokens required to represent text. To overcome this issue, we present a methodology named Cabrita, which, as our research demonstrates, successfully addresses the performance and efficient tokenization problem, all at an affordable cost. We believe that this methodology can be applied to any transformer-like architecture model. To validate the study, we conducted continuous pre-training exclusively using Portuguese text on a 3-billion-parameter model known as OpenLLaMA, resulting in a model named openCabrita 3B. The openCabrita 3B also features a new tokenizer that results in a significant reduction in the number of tokens required to represent the text. In our assessment, for few-shot learning tasks, we achieved similar results with this 3B model compared to a traditional continuous pre-training approach as well as to 7B models English pre-trained models.