For readability assessment, traditional methods mainly employ machine learning classifiers with hundreds of linguistic features. Although the deep learning model has become the prominent approach for almost all NLP tasks, it is less explored for readability assessment. In this paper, we propose a BERT-based model with feature projection and length-balanced loss (BERT-FP-LBL) for readability assessment. Specially, we present a new difficulty knowledge guided semi-supervised method to extract topic features to complement the traditional linguistic features. From the linguistic features, we employ projection filtering to extract orthogonal features to supplement BERT representations. Furthermore, we design a new length-balanced loss to handle the greatly varying length distribution of data. Our model achieves state-of-the-art performances on two English benchmark datasets and one dataset of Chinese textbooks, and also achieves the near-perfect accuracy of 99\% on one English dataset. Moreover, our proposed model obtains comparable results with human experts in consistency test.

Deep Neural Networks (DNNs) are being adopted as components in software systems. Creating and specializing DNNs from scratch has grown increasingly difficult as state-of-the-art architectures grow more complex. Following the path of traditional software engineering, machine learning engineers have begun to reuse large-scale pre-trained models (PTMs) and fine-tune these models for downstream tasks. Prior works have studied reuse practices for traditional software packages to guide software engineers towards better package maintenance and dependency management. We lack a similar foundation of knowledge to guide behaviors in pre-trained model ecosystems. In this work, we present the first empirical investigation of PTM reuse. We interviewed 12 practitioners from the most popular PTM ecosystem, Hugging Face, to learn the practices and challenges of PTM reuse. From this data, we model the decision-making process for PTM reuse. Based on the identified practices, we describe useful attributes for model reuse, including provenance, reproducibility, and portability. Three challenges for PTM reuse are missing attributes, discrepancies between claimed and actual performance, and model risks. We substantiate these identified challenges with systematic measurements in the Hugging Face ecosystem. Our work informs future directions on optimizing deep learning ecosystems by automated measuring useful attributes and potential attacks, and envision future research on infrastructure and standardization for model registries.

These are some of the important things to understand as we wrap our brains around what this is and how to navigate it in the classroom. I'll do plenty of my own human thinking and reasoning, but for the purposes of information, I'm going to let ChatGPT give us working definitions to use: I am an artificial intelligence assistant trained by OpenAI to help answer questions and provide information on a wide variety of topics. I am not a physical being, but rather a program that is designed to process and generate text based on the input I receive. My primary function is to assist users like you by providing information and answering questions to the best of my ability. I have been designed to have a conversational style and can understand and respond to natural language input.

The growing popularity of artificial intelligence (AI) programmes, which have shown themselves increasingly capable in recent months of generating images, videos, music, computer programming code and even texts of all kinds in a matter of seconds, producing seemingly appropriate and coherent results, in many instances – and in many others, not – is arousing fascination and concern all over the world, especially among artists and creators. What the AI tools of today can do is, at times, so spectacular and convincing that it is hard not to think it must be the work of a conscious being that comprehends what is being asked of it and understands what it produces in response. This is clearly not the case, but for the public at large it suddenly seems like we are witnessing the sudden emergence of revolutionary technology, full of potential and promise but also perils that could transform our world. This day may come, but it is further away than the flurry of expectation may lead us to think. What has happened in recent months, above all, is that the current technology, quite widespread and known to all researchers who had hitherto been experimenting with it behind closed doors, has suddenly started to see the light of day, not only with a view to introducing it to the public, arousing interest and attracting investors, but also so that the programmes could benefit from interacting with people and be'trained' by millions of requests and users at the same time, a massive amount of activity and information that no company could otherwise secure for their AIs.

The Microsoft Azure Cognitive Speech Services platform is a comprehensive collection of technologies and services aimed at accelerating the incorporation of speech into applications and amplifying differentiation to the market as a result. Among the services available are Speech to Text, Text to Speech, custom neural voice (CNV) Conversation Transcription Service, Speaker Recognition, Speech Translation, Speech SDK, and Speech Device Development Kit (DDK). AI for education is an emerging technology that has the potential to revolutionize the way we teach and learn languages. One of the most important aspects of language learning is the ability to pronounce words accurately, and this is…

Synthesizing realistic co-speech gestures is an important and yet unsolved problem for creating believable motions that can drive a humanoid robot to interact and communicate with human users. Such capability will improve the impressions of the robots by human users and will find applications in education, training, and medical services. One challenge in learning the co-speech gesture model is that there may be multiple viable gesture motions for the same speech utterance. The deterministic regression methods can not resolve the conflicting samples and may produce over-smoothed or damped motions. We proposed a two-stage model to address this uncertainty issue in gesture synthesis by modeling the gesture segments as discrete latent codes. Our method utilizes RQ-VAE in the first stage to learn a discrete codebook consisting of gesture tokens from training data. In the second stage, a two-level autoregressive transformer model is used to learn the prior distribution of residual codes conditioned on input speech context. Since the inference is formulated as token sampling, multiple gesture sequences could be generated given the same speech input using top-k sampling. The quantitative results and the user study showed the proposed method outperforms the previous methods and is able to generate realistic and diverse gesture motions.

Many graph neural networks have been developed to learn graph representations in either Euclidean or hyperbolic space, with all nodes' representations embedded in a single space. However, a graph can have hyperbolic and Euclidean geometries at different regions of the graph. Thus, it is sub-optimal to indifferently embed an entire graph into a single space. In this paper, we explore and analyze two notions of local hyperbolicity, describing the underlying local geometry: geometric (Gromov) and model-based, to determine the preferred space of embedding for each node. The two hyperbolicities' distributions are aligned using the Wasserstein metric such that the calculated geometric hyperbolicity guides the choice of the learned model hyperbolicity. As such our model Joint Space Graph Neural Network (JSGNN) can leverage both Euclidean and hyperbolic spaces during learning by allowing node-specific geometry space selection. We evaluate our model on both node classification and link prediction tasks and observe promising performance compared to baseline models.

ML platforms help enable intelligent data-driven applications and maintain them with limited engineering effort. Upon sufficiently broad adoption, such platforms reach economies of scale that bring greater component reuse while improving efficiency of system development and maintenance. For an end-to-end ML platform with broad adoption, scaling relies on pervasive ML automation and system integration to reach the quality we term self-serve that we define with ten requirements and six optional capabilities. With this in mind, we identify long-term goals for platform development, discuss related tradeoffs and future work. Our reasoning is illustrated on two commercially-deployed end-to-end ML platforms that host hundreds of real-time use cases -- one general-purpose and one specialized.

Concept learning deals with learning description logic concepts from a background knowledge and input examples. The goal is to learn a concept that covers all positive examples, while not covering any negative examples. This non-trivial task is often formulated as a search problem within an infinite quasi-ordered concept space. Although state-of-the-art models have been successfully applied to tackle this problem, their large-scale applications have been severely hindered due to their excessive exploration incurring impractical runtimes. Here, we propose a remedy for this limitation. We reformulate the learning problem as a multi-label classification problem and propose a neural embedding model (NERO) that learns permutation-invariant embeddings for sets of examples tailored towards predicting $F_1$ scores of pre-selected description logic concepts. By ranking such concepts in descending order of predicted scores, a possible goal concept can be detected within few retrieval operations, i.e., no excessive exploration. Importantly, top-ranked concepts can be used to start the search procedure of state-of-the-art symbolic models in multiple advantageous regions of a concept space, rather than starting it in the most general concept $\top$. Our experiments on 5 benchmark datasets with 770 learning problems firmly suggest that NERO significantly (p-value <1%) outperforms the state-of-the-art models in terms of $F_1$ score, the number of explored concepts, and the total runtime. We provide an open-source implementation of our approach.

Imitation learning aims to mimic the behavior of experts without explicit reward signals. Passive imitation learning methods which use static expert datasets typically suffer from compounding error, low sample efficiency, and high hyper-parameter sensitivity. In contrast, active imitation learning methods solicit expert interventions to address the limitations. However, recent active imitation learning methods are designed based on human intuitions or empirical experience without theoretical guarantee. In this paper, we propose a novel active imitation learning framework based on a teacher-student interaction model, in which the teacher's goal is to identify the best teaching behavior and actively affect the student's learning process. By solving the optimization objective of this framework, we propose a practical implementation, naming it AdapMen. Theoretical analysis shows that AdapMen can improve the error bound and avoid compounding error under mild conditions. Experiments on the MetaDrive benchmark and Atari 2600 games validate our theoretical analysis and show that our method achieves near-expert performance with much less expert involvement and total sampling steps than previous methods. The code is available at https://github.com/liuxhym/AdapMen.