Structure-agnostic causal inference studies how well one can estimate a treatment effect given black-box machine learning estimates of nuisance functions (like the impact of confounders on treatment and outcomes). Here, we find that the answer depends in a surprising way on the distribution of the treatment noise. Focusing on the partially linear model of \citet{robinson1988root}, we first show that the widely adopted double machine learning (DML) estimator is minimax rate-optimal for Gaussian treatment noise, resolving an open problem of \citet{mackey2018orthogonal}. Meanwhile, for independent non-Gaussian treatment noise, we show that DML is always suboptimal by constructing new practical procedures with higher-order robustness to nuisance errors. These \emph{ACE} procedures use structure-agnostic cumulant estimators to achieve $r$-th order insensitivity to nuisance errors whenever the $(r+1)$-st treatment cumulant is non-zero. We complement these core results with novel minimax guarantees for binary treatments in the partially linear model. Finally, using synthetic demand estimation experiments, we demonstrate the practical benefits of our higher-order robust estimators.

Large Language Models (LLMs) often encounter conflicts between their learned, internal (parametric knowledge, PK) and external knowledge provided during inference (contextual knowledge, CK). Understanding how LLMs models prioritize one knowledge source over the other remains a challenge. In this paper, we propose a novel probing framework to explore the mechanisms governing the selection between PK and CK in LLMs. Using controlled prompts designed to contradict the model's PK, we demonstrate that specific model activations are indicative of the knowledge source employed. We evaluate this framework on various LLMs of different sizes and demonstrate that mid-layer activations, particularly those related to relations in the input, are crucial in predicting knowledge source selection, paving the way for more reliable models capable of handling knowledge conflicts effectively.

Large language models (LLMs) often exhibit hallucinations due to incorrect or outdated knowledge. Hence, model editing methods have emerged to enable targeted knowledge updates. To achieve this, a prevailing paradigm is the locating-then-editing approach, which first locates influential parameters and then edits them by introducing a perturbation. While effective, current studies have demonstrated that this perturbation inevitably disrupt the originally preserved knowledge within LLMs, especially in sequential editing scenarios. To address this, we introduce AlphaEdit, a novel solution that projects perturbation onto the null space of the preserved knowledge before applying it to the parameters. We theoretically prove that this projection ensures the output of post-edited LLMs remains unchanged when queried about the preserved knowledge, thereby mitigating the issue of disruption. Extensive experiments on various LLMs, including LLaMA3, GPT2-XL, and GPT-J, show that AlphaEdit boosts the performance of most locating-then-editing methods by an average of 36.4% with a single line of additional code for projection solely. Our code is available at: https://github.com/jianghoucheng/AlphaEdit.

Authorship profiling is the process of identifying an author's characteristics based on their writings. This centuries old problem has become more intriguing especially with recent developments in Natural Language Processing (NLP). In this paper, we introduce a corpus of short texts in the Romanian language, annotated with certain author characteristic keywords; to our knowledge, the first of its kind. In order to do this, we exploit a social media platform called Reddit. We leverage its thematic community-based structure (subreddits structure), which offers information about the author's background. We infer an user's demographic and some broad personal traits, such as age category, employment status, interests, and social orientation based on the subreddit and other cues. We thus obtain a 23k+ samples corpus, extracted from 100+ Romanian subreddits. We analyse our dataset, and finally, we fine-tune and evaluate Large Language Models (LLMs) to prove baselines capabilities for authorship profiling using the corpus, indicating the need for further research in the field. We publicly release all our resources.

Training deep models for time series forecasting is a critical task with an inherent challenge of time complexity. While current methods generally ensure linear time complexity, our observations on temporal redundancy show that high-level features are learned 98.44\% slower than low-level features. To address this issue, we introduce a new exponentially weighted stochastic gradient descent algorithm designed to achieve constant time complexity in deep learning models. We prove that the theoretical complexity of this learning method is constant. Evaluation of this method on Kernel U-Net (K-U-Net) on synthetic datasets shows a significant reduction in complexity while improving the accuracy of the test set.

This paper introduces a novel two-stage framework for accurate Egyptian Vehicle License Plate Recognition (EVLPR). The first stage employs image processing techniques to reliably localize license plates, while the second stage utilizes a custom-designed deep learning model for robust Arabic character recognition. The proposed system achieves a remarkable 99.3% accuracy on a diverse dataset, surpassing existing approaches. Its potential applications extend to intelligent traffic management, including traffic violation detection and parking optimization. Future research will focus on enhancing the system's capabilities through architectural refinements, expanded datasets, and addressing system dependencies.

Estimation of football players' skills is one of the key tasks in sports analytics. This paper introduces multiple extensions to a widely used model, expected possession value (EPV), to address some key challenges such as selection problem. First, we assign greater weights to events occurring immediately prior to the shot rather than those preceding them (decay effect). Second, our model incorporates possession risk more accurately by considering the decay effect and effective playing time. Third, we integrate the assessment of individual player ability to win aerial and ground duels. Using the extended EPV model, we predict this metric for various football players for the upcoming season, particularly taking into account the strength of their opponents.

This paper addresses the ethical challenges of Artificial Intelligence in Neural Machine Translation (NMT) systems, emphasizing the imperative for developers to ensure fairness and cultural sensitivity. We investigate the ethical competence of AI models in NMT, examining the Ethical considerations at each stage of NMT development, including data handling, privacy, data ownership, and consent. We identify and address ethical issues through empirical studies. These include employing Transformer models for Luganda-English translations and enhancing efficiency with sentence mini-batching. And complementary studies that refine data labeling techniques and fine-tune BERT and Longformer models for analyzing Luganda and English social media content. Our second approach is a literature review from databases such as Google Scholar and platforms like GitHub. Additionally, the paper probes the distribution of responsibility between AI systems and humans, underscoring the essential role of human oversight in upholding NMT ethical standards. Incorporating a biblical perspective, we discuss the societal impact of NMT and the broader ethical responsibilities of developers, positing them as stewards accountable for the societal repercussions of their creations.

Compiling large datasets from published resources, such as archaeological find catalogues presents fundamental challenges: identifying relevant content and manually recording it is a time-consuming, repetitive and error-prone task. For the data to be useful, it must be of comparable quality and adhere to the same recording standards, which is hardly ever the case in archaeology. Here, we present a new data collection method exploiting recent advances in Artificial Intelligence. Our software uses an object detection neural network combined with further classification networks to speed up, automate, and standardise data collection from legacy resources, such as archaeological drawings and photographs in large unsorted PDF files. The AI-assisted workflow detects common objects found in archaeological catalogues, such as graves, skeletons, ceramics, ornaments, stone tools and maps, and spatially relates and analyses these objects on the page to extract real-life attributes, such as the size and orientation of a grave based on the north arrow and the scale. A graphical interface allows for and assists with manual validation. We demonstrate the benefits of this approach by collecting a range of shapes and numerical attributes from richly-illustrated archaeological catalogues, and benchmark it in a real-world experiment with ten users.

Dialect identification is a critical task in speech processing and language technology, enhancing various applications such as speech recognition, speaker verification, and many others. While most research studies have been dedicated to dialect identification in widely spoken languages, limited attention has been given to dialect identification in low-resource languages, such as Romanian. To address this research gap, we introduce RoDia, the first dataset for Romanian dialect identification from speech. The RoDia dataset includes a varied compilation of speech samples from five distinct regions of Romania, covering both urban and rural environments, totaling 2 hours of manually annotated speech data. Along with our dataset, we introduce a set of competitive models to be used as baselines for future research. The top scoring model achieves a macro F1 score of 59.83% and a micro F1 score of 62.08%, indicating that the task is challenging. We thus believe that RoDia is a valuable resource that will stimulate research aiming to address the challenges of Romanian dialect identification. We publicly release our dataset and code at https://github.com/codrut2/RoDia.