Recent advancements in Large Language Models (LLMs) harness linguistic associations in vast natural language data for practical applications. However, their ability to understand the physical world using only language data remains a question. After reviewing existing protocols, we explore this question using a novel and tightly controlled reasoning test (ART) and compare human norms against versions of GPT-3. Our findings highlight the categories of common-sense relations models that could learn directly from data and areas of weakness. GPT-3 offers evidence for verbal reasoning on a par with human subjects for several relations including Synonymy, Antonymy, and Default inheritance, Without reinforcement learning from human judgements, it appears GPT-3 performs at the lower end of the reference interval for Has-part and Contained-in. Weaknesses were observed also in affordance characteristics through Necessary-quality, Order-of-size and Order-of-intensity. Combining LLMs with symbolic world grounding is a promising direction to address associative learning.

Lifelong graph learning deals with the problem of continually adapting graph neural network (GNN) models to changes in evolving graphs. We address two critical challenges of lifelong graph learning in this work: dealing with new classes and tackling imbalanced class distributions. The combination of these two challenges is particularly relevant since newly emerging classes typically resemble only a tiny fraction of the data, adding to the already skewed class distribution. We make several contributions: First, we show that the amount of unlabeled data does not influence the results, which is an essential prerequisite for lifelong learning on a sequence of tasks. Second, we experiment with different label rates and show that our methods can perform well with only a tiny fraction of annotated nodes. Third, we propose the gDOC method to detect new classes under the constraint of having an imbalanced class distribution. The critical ingredient is a weighted binary cross-entropy loss function to account for the class imbalance. Moreover, we demonstrate combinations of gDOC with various base GNN models such as GraphSAGE, Simplified Graph Convolution, and Graph Attention Networks. Lastly, our k-neighborhood time difference measure provably normalizes the temporal changes across different graph datasets. With extensive experimentation, we find that the proposed gDOC method is consistently better than a naive adaption of DOC to graphs. Specifically, in experiments using the smallest history size, the out-of-distribution detection score of gDOC is 0.09 compared to 0.01 for DOC. Furthermore, gDOC achieves an Open-F1 score, a combined measure of in-distribution classification and out-of-distribution detection, of 0.33 compared to 0.25 of DOC (32% increase).

Many recent machine learning tasks focus to develop models that can generalize to unseen distributions. Domain generalization (DG) has become one of the key topics in various fields. Several literatures show that DG can be arbitrarily hard without exploiting target domain information. To address this issue, test-time adaptive (TTA) methods are proposed. Existing TTA methods require offline target data or extra sophisticated optimization procedures during the inference stage. In this work, we adopt Non-Parametric Classifier to perform the test-time Adaptation (AdaNPC). In particular, we construct a memory that contains the feature and label pairs from training domains. During inference, given a test instance, AdaNPC first recalls K closed samples from the memory to vote for the prediction, and then the test feature and predicted label are added to the memory. In this way, the sample distribution in the memory can be gradually changed from the training distribution towards the test distribution with very little extra computation cost. We theoretically justify the rationality behind the proposed method. Besides, we test our model on extensive numerical experiments. AdaNPC significantly outperforms competitive baselines on various DG benchmarks. In particular, when the adaptation target is a series of domains, the adaptation accuracy of AdaNPC is 50% higher than advanced TTA methods. The code is available at https://github.com/yfzhang114/AdaNPC.

Knowledge distillation is the technique of compressing a larger neural network, known as the teacher, into a smaller neural network, known as the student, while still trying to maintain the performance of the larger neural network as much as possible. Existing methods of knowledge distillation are mostly applicable for classification tasks. Many of them also require access to the data used to train the teacher model. To address the problem of knowledge distillation for regression tasks under the absence of original training data, previous work has proposed a data-free knowledge distillation method where synthetic data are generated using a generator model trained adversarially against the student model. These synthetic data and their labels predicted by the teacher model are then used to train the student model. In this study, we investigate the behavior of various synthetic data generation methods and propose a new synthetic data generation strategy that directly optimizes for a large but bounded difference between the student and teacher model. Our results on benchmark and case study experiments demonstrate that the proposed strategy allows the student model to learn better and emulate the performance of the teacher model more closely.

Prof Verena Rieser, a computer scientist at Heriot-Watt University who has been working in the AI field for two decades, says her own students are "using it in very creative ways" - but chatbots are still in the early stages of development and "can be used to generate misinformation at [a] scale which is obviously very concerning" when it comes to education.

Relation extraction (RE) is the core NLP task of inferring semantic relationships between entities from text. Standard supervised RE techniques entail training modules to tag tokens comprising entity spans and then predict the relationship between them. Recent work has instead treated the problem as a \emph{sequence-to-sequence} task, linearizing relations between entities as target strings to be generated conditioned on the input. Here we push the limits of this approach, using larger language models (GPT-3 and Flan-T5 large) than considered in prior work and evaluating their performance on standard RE tasks under varying levels of supervision. We address issues inherent to evaluating generative approaches to RE by doing human evaluations, in lieu of relying on exact matching. Under this refined evaluation, we find that: (1) Few-shot prompting with GPT-3 achieves near SOTA performance, i.e., roughly equivalent to existing fully supervised models; (2) Flan-T5 is not as capable in the few-shot setting, but supervising and fine-tuning it with Chain-of-Thought (CoT) style explanations (generated via GPT-3) yields SOTA results. We release this model as a new baseline for RE tasks.

The design made it possible to make powerful language models like term "Generative AI" is used to describe a subset of AI models OpenAI's GPT series, which included GPT-2 and GPT-3, that can generate new information by discovering relevant which were the versions that came before ChatGPT [6]. The trends and patterns in already collected information. These GPT-3.5 architecture is the basis for ChatGPT; it is an models may produce work in a wide range of media, from improved version of OpenAI's GPT-3 model. Even though written to visual to audio [2]. To analyse, comprehend, and GPT-3.5 has fewer variables, nevertheless produces excellent produce material that accurately imitates human-generated results in many areas of NLP, such as language understanding, outcomes, Generative AI models depend on deep learning text generation, and machine translation [6]. ChatGPT was approaches and neural networks. OpenAI's ChatGPT is one trained on a massive body of text data and fine-tuned on the such AI model that has quickly become a popular and versatile goal of creating conversational replies, allowing it to create resource for a number of different industries. Its humanoid text responses to user inquiries that are strangely similar to those of generation is made possible by its foundation in the Generative a person.

Automated feedback as students answer open-ended math questions has significant potential in improving learning outcomes at large scale. A key part of automated feedback systems is an error classification component, which identifies student errors and enables appropriate, predefined feedback to be deployed. Most existing approaches to error classification use a rule-based method, which has limited capacity to generalize. Existing data-driven methods avoid these limitations but specifically require mathematical expressions in student responses to be parsed into syntax trees. This requirement is itself a limitation, since student responses are not always syntactically valid and cannot be converted into trees. In this work, we introduce a flexible method for error classification using pre-trained large language models. We demonstrate that our method can outperform existing methods in algebra error classification, and is able to classify a larger set of student responses. Additionally, we analyze common classification errors made by our method and discuss limitations of automated error classification.

Traditional learning problems typically assume data distributions to be static. For applications such as face recognition, this is largely true and designing algorithms under such an assumption in general does not impact learning efficacy. This, however, is not true in many other domains. In some cases, there may be a natural evolution and shift in the distribution, e.g., in weather and climate data, in which case new data need to be acquired periodically and the algorithm re-trained to remain up to date. In other cases, the distribution shift is the result of the very learning outcome, when individuals respond to the algorithmic decisions they are subjected to. For instance, when users with certain accents perceive larger-than-acceptable errors from a speech recognition software and therefore stop using it, this can directly impact the type of speech samples collected by the software used for training the next generation of the product. Another example is "gaming the algorithm", where users through honest or dishonest means attempt to improve critical features so as to obtain a favorable decision by the algorithm (e.g., in loan approvals or job applications). This again can directly lead to the distributional change in features and label that the algorithm relies on for decision making.

Optimal Transport (OT) is a mathematical framework that first emerged in the eighteenth century and has led to a plethora of methods for answering many theoretical and applied questions. The last decade is a witness of the remarkable contributions of this classical optimization problem to machine learning. This paper is about where and how optimal transport is used in machine learning with a focus on the question of salable optimal transport. We provide a comprehensive survey of optimal transport while ensuring an accessible presentation as permitted by the nature of the topic and the context. First, we explain optimal transport background and introduce different flavors (i.e. mathematical formulations), properties, and notable applications. We then address the fundamental question of how to scale optimal transport to cope with the current demands of big and high dimensional data. We conduct a systematic analysis of the methods used in the literature for scaling OT and present the findings in a unified taxonomy. We conclude with presenting some open challenges and discussing potential future research directions. A live repository of related OT research papers is maintained in https://github.com/abdelwahed/OT_for_big_data.git.