While many speakers of low-resource languages regularly code-switch between their languages and other regional languages or English, datasets of codeswitched speech are too small to train bespoke acoustic models from scratch or do language model rescoring. Here we propose finetuning self-supervised speech representations such as wav2vec 2.0 XLSR to recognize code-switched data. We find that finetuning self-supervised multilingual representations and augmenting them with n-gram language models trained from transcripts reduces absolute word error rates by up to 20% compared to baselines of hybrid models trained from scratch on code-switched data. Our findings suggest that in circumstances with limited training data finetuning self-supervised representations is a better performing and viable solution.

Large language models (LLMs) have demonstrated remarkable capabilities across a broad spectrum of tasks. They have attracted significant attention and been deployed in numerous downstream applications. Nevertheless, akin to a double-edged sword, LLMs also present potential risks. They could suffer from private data leaks or yield inappropriate, harmful, or misleading content. Additionally, the rapid progress of LLMs raises concerns about the potential emergence of superintelligent systems without adequate safeguards. To effectively capitalize on LLM capacities as well as ensure their safe and beneficial development, it is critical to conduct a rigorous and comprehensive evaluation of LLMs. This survey endeavors to offer a panoramic perspective on the evaluation of LLMs. We categorize the evaluation of LLMs into three major groups: knowledge and capability evaluation, alignment evaluation and safety evaluation. In addition to the comprehensive review on the evaluation methodologies and benchmarks on these three aspects, we collate a compendium of evaluations pertaining to LLMs' performance in specialized domains, and discuss the construction of comprehensive evaluation platforms that cover LLM evaluations on capabilities, alignment, safety, and applicability. We hope that this comprehensive overview will stimulate further research interests in the evaluation of LLMs, with the ultimate goal of making evaluation serve as a cornerstone in guiding the responsible development of LLMs. We envision that this will channel their evolution into a direction that maximizes societal benefit while minimizing potential risks. A curated list of related papers has been publicly available at https://github.com/tjunlp-lab/Awesome-LLMs-Evaluation-Papers.

What is the relationship between model architecture and the ability to perform in-context learning? In this empirical study, we take the first steps toward answering this question. We evaluate twelve model architectures capable of causal language modeling across a suite of synthetic in-context learning tasks. These selected architectures represent a broad range of paradigms, including recurrent and convolution-based neural networks, transformers, state-space model inspired, and other emerging attention alternatives. We discover that all the considered architectures can perform in-context learning under a wider range of conditions than previously documented. Additionally, we observe stark differences in statistical efficiency and consistency by varying context length and task difficulty. We also measure each architecture's predisposition towards in-context learning when presented with alternative routes for task resolution. Finally, and somewhat surprisingly, we find that several attention alternatives are more robust in-context learners than transformers. Given that such approaches have constant-sized memory footprints at inference time, this result opens the possibility of scaling up in-context learning to accommodate vastly larger numbers of in-context examples. In-context learning (ICL) refers to the ability to learn new tasks at inference time, using only inputoutput pair exemplars as guidance. Radford et al. (2019) demonstrate early signs of this ability in GPT-2, a causal transformer (Vaswani et al., 2017). ICL was further popularized by GPT-3 (Brown et al., 2020), a large language model with the same architectural foundation but augmented with greater capacity and trained on large-scale data. By simply adjusting a natural language prompt, it was shown that GPT-3 could adapt to new tasks, such as translation and question answering, without updating any of its parameters.

Treatment pathways are step-by-step plans outlining the recommended medical care for specific diseases; they get revised when different treatments are found to improve patient outcomes. Examining health records is an important part of this revision process, but inferring patients' actual treatments from health data is challenging due to complex event-coding schemes and the absence of pathway-related annotations. This study aims to infer the actual treatment steps for a particular patient group from administrative health records (AHR) - a common form of tabular healthcare data - and address several technique- and methodology-based gaps in treatment pathway-inference research. We introduce Defrag, a method for examining AHRs to infer the real-world treatment steps for a particular patient group. Defrag learns the semantic and temporal meaning of healthcare event sequences, allowing it to reliably infer treatment steps from complex healthcare data. To our knowledge, Defrag is the first pathway-inference method to utilise a neural network (NN), an approach made possible by a novel, self-supervised learning objective. We also developed a testing and validation framework for pathway inference, which we use to characterise and evaluate Defrag's pathway inference ability and compare against baselines. We demonstrate Defrag's effectiveness by identifying best-practice pathway fragments for breast cancer, lung cancer, and melanoma in public healthcare records. Additionally, we use synthetic data experiments to demonstrate the characteristics of the Defrag method, and to compare Defrag to several baselines where it significantly outperforms non-NN-based methods. Defrag significantly outperforms several existing pathway-inference methods and offers an innovative and effective approach for inferring treatment pathways from AHRs. Open-source code is provided to encourage further research in this area.

This work uses visual knowledge discovery in parallel coordinates to advance methods of interpretable machine learning. The graphic data representation in parallel coordinates made the concepts of hypercubes and hyperblocks (HBs) simple to understand for end users. It is suggested to use mixed and pure hyperblocks in the proposed data classifier algorithm Hyper. It is shown that Hyper models generalize decision trees. The algorithm is presented in several settings and options to discover interactively or automatically overlapping or non-overlapping hyperblocks. Additionally, the use of hyperblocks in conjunction with language descriptions of visual patterns is demonstrated. The benchmark data from the UCI ML repository were used to evaluate the Hyper algorithm. It enabled the discovery of mixed and pure HBs evaluated using 10-fold cross validation. Connections among hyperblocks, dimension reduction and visualization have been established. The capability of end users to find and observe hyperblocks, as well as the ability of side-by-side visualizations to make patterns evident, are among major advantages ofhyperblock technology and the Hyper algorithm. A new method to visualize incomplete n-D data with missing values is proposed, while the traditional parallel coordinates do not support it. The ability of HBs to better prevent both overgeneralization and overfitting of data over decision trees is demonstrated as another benefit of the hyperblocks. The features of VisCanvas 2.0 software tool that implements Hyper technology are presented.

Generalized Additive Models (GAMs) have recently experienced a resurgence in popularity due to their interpretability, which arises from expressing the target value as a sum of non-linear transformations of the features. Despite the current enthusiasm for GAMs, their susceptibility to concurvity - i.e., (possibly non-linear) dependencies between the features - has hitherto been largely overlooked. Here, we demonstrate how concurvity can severly impair the interpretability of GAMs and propose a remedy: a conceptually simple, yet effective regularizer which penalizes pairwise correlations of the non-linearly transformed feature variables. This procedure is applicable to any differentiable additive model, such as Neural Additive Models or NeuralProphet, and enhances interpretability by eliminating ambiguities due to self-canceling feature contributions. We validate the effectiveness of our regularizer in experiments on synthetic as well as real-world datasets for time-series and tabular data. Our experiments show that concurvity in GAMs can be reduced without significantly compromising prediction quality, improving interpretability and reducing variance in the feature importances.

Over the past decade, Time Series Classification (TSC) has gained an increasing attention. While various methods were explored, deep learning - particularly through Convolutional Neural Networks (CNNs)-stands out as an effective approach. However, due to the limited availability of training data, defining a foundation model for TSC that overcomes the overfitting problem is still a challenging task. The UCR archive, encompassing a wide spectrum of datasets ranging from motion recognition to ECG-based heart disease detection, serves as a prime example for exploring this issue in diverse TSC scenarios. In this paper, we address the overfitting challenge by introducing pre-trained domain foundation models. A key aspect of our methodology is a novel pretext task that spans multiple datasets. This task is designed to identify the originating dataset of each time series sample, with the goal of creating flexible convolution filters that can be applied across different datasets. The research process consists of two phases: a pre-training phase where the model acquires general features through the pretext task, and a subsequent fine-tuning phase for specific dataset classifications. Our extensive experiments on the UCR archive demonstrate that this pre-training strategy significantly outperforms the conventional training approach without pre-training. This strategy effectively reduces overfitting in small datasets and provides an efficient route for adapting these models to new datasets, thus advancing the capabilities of deep learning in TSC.

Artificial Intelligence (AI) is one of the most transformative technologies of the 21st century. The extent and scope of future AI capabilities remain a key uncertainty, with widespread disagreement on timelines and potential impacts. As nations and technology companies race toward greater complexity and autonomy in AI systems, there are concerns over the extent of integration and oversight of opaque AI decision processes. This is especially true in the subfield of machine learning (ML), where systems learn to optimize objectives without human assistance. Objectives can be imperfectly specified or executed in an unexpected or potentially harmful way. This becomes more concerning as systems increase in power and autonomy, where an abrupt capability jump could result in unexpected shifts in power dynamics or even catastrophic failures. This study presents a hierarchical complex systems framework to model AI risk and provide a template for alternative futures analysis. Survey data were collected from domain experts in the public and private sectors to classify AI impact and likelihood. The results show increased uncertainty over the powerful AI agent scenario, confidence in multiagent environments, and increased concern over AI alignment failures and influence-seeking behavior.

Temporal information extraction (TIE) has attracted a great deal of interest over the last two decades, leading to the development of a significant number of datasets. Despite its benefits, having access to a large volume of corpora makes it difficult when it comes to benchmark TIE systems. On the one hand, different datasets have different annotation schemes, thus hindering the comparison between competitors across different corpora. On the other hand, the fact that each corpus is commonly disseminated in a different format requires a considerable engineering effort for a researcher/practitioner to develop parsers for all of them. This constraint forces researchers to select a limited amount of datasets to evaluate their systems which consequently limits the comparability of the systems. Yet another obstacle that hinders the comparability of the TIE systems is the evaluation metric employed. While most research works adopt traditional metrics such as precision, recall, and $F_1$, a few others prefer temporal awareness -- a metric tailored to be more comprehensive on the evaluation of temporal systems. Although the reason for the absence of temporal awareness in the evaluation of most systems is not clear, one of the factors that certainly weights this decision is the necessity to implement the temporal closure algorithm in order to compute temporal awareness, which is not straightforward to implement neither is currently easily available. All in all, these problems have limited the fair comparison between approaches and consequently, the development of temporal extraction systems. To mitigate these problems, we have developed tieval, a Python library that provides a concise interface for importing different corpora and facilitates system evaluation. In this paper, we present the first public release of tieval and highlight its most relevant features.

An increasing number of studies use gender information to understand phenomena such as gender bias, inequity in access and participation, or the impact of the Covid pandemic response. Unfortunately, most datasets do not include self-reported gender information, making it necessary for researchers to infer gender from other information, such as names or names and country information. An important limitation of these tools is that they fail to appropriately capture the fact that gender exists on a non-binary scale, however, it remains important to evaluate and compare how well these tools perform in a variety of contexts. In this paper, we compare the performance of a generative Artificial Intelligence (AI) tool ChatGPT with three commercially available list-based and machine learning-based gender inference tools (Namsor, Gender-API, and genderize.io) on a unique dataset. Specifically, we use a large Olympic athlete dataset and report how variations in the input (e.g., first name and first and last name, with and without country information) impact the accuracy of their predictions. We report results for the full set, as well as for the subsets: medal versus non-medal winners, athletes from the largest English-speaking countries, and athletes from East Asia. On these sets, we find that Namsor is the best traditional commercially available tool. However, ChatGPT performs at least as well as Namsor and often outperforms it, especially for the female sample when country and/or last name information is available. All tools perform better on medalists versus non-medalists and on names from English-speaking countries. Although not designed for this purpose, ChatGPT may be a cost-effective tool for gender prediction. In the future, it might even be possible for ChatGPT or other large scale language models to better identify self-reported gender rather than report gender on a binary scale.