What will the market size be in 2027 and what will the growth rate be? What are the key market trends? What is driving this market? What are the challenges to market growth? Who are the key vendors in this market space?

In this paper, we study the importance of context in predicting the citation worthiness of sentences in scholarly articles. We formulate this problem as a sequence labeling task solved using a hierarchical BiLSTM model. We contribute a new benchmark dataset containing over two million sentences and their corresponding labels. We preserve the sentence order in this dataset and perform document-level train/test splits, which importantly allows incorporating contextual information in the modeling process. We evaluate the proposed approach on three benchmark datasets. Our results quantify the benefits of using context and contextual embeddings for citation worthiness. Lastly, through error analysis, we provide insights into cases where context plays an essential role in predicting citation worthiness.

Traditionally Genetic Algorithm has been used for optimization of unimodal and multimodal functions. Earlier researchers worked with constant probabilities of GA control operators like crossover, mutation etc. for tuning the optimization in specific domains. Recent advancements in this field witnessed adaptive approach in probability determination. In Adaptive mutation primarily poor individuals are utilized to explore state space, so mutation probability is usually generated proportionally to the difference between fitness of best chromosome and itself (fMAX - f). However, this approach is susceptible to nature of fitness distribution during optimization. This paper presents an alternate approach of mutation probability generation using chromosome rank to avoid any susceptibility to fitness distribution. Experiments are done to compare results of simple genetic algorithm (SGA) with constant mutation probability and adaptive approaches within a limited resource constraint for unimodal, multimodal functions and Travelling Salesman Problem (TSP). Measurements are done for average best fitness, number of generations evolved and percentage of global optimum achievements out of several trials. The results demonstrate that the rank-based adaptive mutation approach is superior to fitness-based adaptive approach as well as SGA in a multimodal problem space.

An interpretable system for complex, open-domain reasoning needs an interpretable meaning representation. Natural language is an excellent candidate -- it is both extremely expressive and easy for humans to understand. However, manipulating natural language statements in logically consistent ways is hard. Models have to be precise, yet robust enough to handle variation in how information is expressed. In this paper, we describe ParaPattern, a method for building models to generate logical transformations of diverse natural language inputs without direct human supervision. We use a BART-based model (Lewis et al., 2020) to generate the result of applying a particular logical operation to one or more premise statements. Crucially, we have a largely automated pipeline for scraping and constructing suitable training examples from Wikipedia, which are then paraphrased to give our models the ability to handle lexical variation. We evaluate our models using targeted contrast sets as well as out-of-domain sentence compositions from the QASC dataset (Khot et al., 2020). Our results demonstrate that our operation models are both accurate and flexible.

IEEE 802.11p standard defines wireless technology protocols that enable vehicular transportation and manage traffic efficiency. A major challenge in the development of this technology is ensuring communication reliability in highly dynamic vehicular environments, where the wireless communication channels are doubly selective, thus making channel estimation and tracking a relevant problem to investigate. In this paper, a novel deep learning (DL)-based weighted interpolation estimator is proposed to accurately estimate vehicular channels especially in high mobility scenarios. The proposed estimator is based on modifying the pilot allocation of the IEEE 802.11p standard so that more transmission data rates are achieved. Extensive numerical experiments demonstrate that the developed estimator significantly outperforms the recently proposed DL-based frame-by-frame estimators in different vehicular scenarios, while substantially reducing the overall computational complexity.

Event schemas are structured knowledge sources defining typical real-world scenarios (e.g., going to an airport). We present a framework for efficient human-in-the-loop construction of a schema library, based on a novel mechanism for schema induction and a wellcrafted interface that allows non-experts to "program" complex event structures. Associated with this work we release a machine readable resource (schema library) of 232 detailed event schemas, each of which describe a distinct typical scenario in terms of its relevant sub-event structure (what happens in the scenario), participants (who plays a role in the scenario), fine-grained typing of each participant, and the implied relational constraints Figure 1: An example schema from our schema library, between them. Our custom annotation interface, induced from a skeleton mined by Causal ARM (Section SchemaBlocks, and the event schemas 3) and fully fleshed out by an annotator using our are available online.

When primed with only a handful of training samples, very large pretrained language models such as GPT-3, have shown competitive results when compared to fully-supervised fine-tuned large pretrained language models. We demonstrate that the order in which the samples are provided can be the difference between near state-of-the-art and random guess performance: Essentially some permutations are "fantastic" and some not. We analyse this phenomenon in detail, establishing that: it is present across model sizes (even for the largest current models), it is not related to a specific subset of samples, and that a given good permutation for one model is not transferable to another. While one could use a development set to determine which permutations are performant, this would deviate from the few-shot setting as it requires additional annotated data. Instead, we use the generative nature of the language models to construct an artificial development set and based on entropy statistics of the candidate permutations from this set we identify performant prompts. Our method improves upon GPT-family models by on average 13% relative across eleven different established text classification tasks.

As language models are trained on ever more text, researchers are turning to some of the largest corpora available. Unlike most other types of datasets in NLP, large unlabeled text corpora are often presented with minimal documentation, and best practices for documenting them have not been established. In this work we provide the first documentation for the Colossal Clean Crawled Corpus (C4; Raffel et al., 2020), a dataset created by applying a set of filters to a single snapshot of Common Crawl. We begin with a high-level summary of the data, including distributions of where the text came from and when it was written. We then give more detailed analysis on salient parts of this data, including the most frequent sources of text (e.g., patents.google.com, which contains a significant percentage of machine translated and/or OCR'd text), the effect that the filters had on the data (they disproportionately remove text in AAE), and evidence that some other benchmark NLP dataset examples are contained in the text. We release a web interface to an interactive, indexed copy of this dataset, encouraging the community to continuously explore and report additional findings.

While day-to-day questions come with a variety of answer types, the current question-answering (QA) literature has failed to adequately address the answer diversity of questions. To this end, we present GooAQ, a large-scale dataset with a variety of answer types. This dataset contains over 5 million questions and 3 million answers collected from Google. GooAQ questions are collected semi-automatically from the Google search engine using its autocomplete feature. This results in naturalistic questions of practical interest that are nonetheless short and expressed using simple language. GooAQ answers are mined from Google's responses to our collected questions, specifically from the answer boxes in the search results. This yields a rich space of answer types, containing both textual answers (short and long) as well as more structured ones such as collections. We benchmarkT5 models on GooAQ and observe that: (a) in line with recent work, LM's strong performance on GooAQ's short-answer questions heavily benefit from annotated data; however, (b) their quality in generating coherent and accurate responses for questions requiring long responses (such as 'how' and 'why' questions) is less reliant on observing annotated data and mainly supported by their pre-training. We release GooAQ to facilitate further research on improving QA with diverse response types.

Europe is already the world's tech privacy cop. Now it might become the AI cop too. Companies using artificial intelligence in the EU could soon be required to get audited first, under new rules set to be proposed by the European Union as soon as next week. The regulations were partly sketched out in an EU white paper last year and aim to ensure the responsible application of AI in high-stakes situations like autonomous driving, remote surgery or predictive policing. Officials want to ensure that such systems are trained on privacy-protecting and diverse data sets.