Text generation is of great importance to many natural language processing applications. However, maximization-based decoding methods (e.g., beam search) of neural language models often lead to degenerate solutions--the generated text is unnatural and contains undesirable repetitions. Existing approaches introduce stochasticity via sampling or modify training objectives to decrease the probabilities of certain tokens (e.g., unlikelihood training). However, they often lead to solutions that lack coherence. In this work, we show that an underlying reason for model degeneration is the anisotropic distribution of token representations. We present a contrastive solution: (i) SimCTG, a contrastive training objective to calibrate the model's representation space, and (ii) a decoding method--contrastive search--to encourage diversity while maintaining coherence in the generated text. Extensive experiments and analyses on three benchmarks from two languages demonstrate that our proposed approach significantly outperforms current state-of-the-art text generation methods as evaluated by both human and automatic metrics.

As free online encyclopedias with massive volumes of content, Wikipedia and Wikidata are key to many Natural Language Processing (NLP) tasks, such as information retrieval, knowledge base building, machine translation, text classification, and text summarization. In this paper, we introduce WikiDes, a novel dataset to generate short descriptions of Wikipedia articles for the problem of text summarization. The dataset consists of over 80k English samples on 6987 topics. We set up a two-phase summarization method - description generation (Phase I) and candidate ranking (Phase II) - as a strong approach that relies on transfer and contrastive learning. For description generation, T5 and BART show their superiority compared to other small-scale pre-trained models. By applying contrastive learning with the diverse input from beam search, the metric fusion-based ranking models outperform the direct description generation models significantly up to 22 ROUGE in topic-exclusive split and topic-independent split. Furthermore, the outcome descriptions in Phase II are supported by human evaluation in over 45.33% chosen compared to 23.66% in Phase I against the gold descriptions. In the aspect of sentiment analysis, the generated descriptions cannot effectively capture all sentiment polarities from paragraphs while doing this task better from the gold descriptions. The automatic generation of new descriptions reduces the human efforts in creating them and enriches Wikidata-based knowledge graphs. Our paper shows a practical impact on Wikipedia and Wikidata since there are thousands of missing descriptions. Finally, we expect WikiDes to be a useful dataset for related works in capturing salient information from short paragraphs. The curated dataset is publicly available at: https://github.com/declare-lab/WikiDes.

In traditional astronomies across the world, groups of stars in the night sky were linked into constellations -- symbolic representations rich in meaning and with practical roles. In some sky cultures, constellations are represented as line (or connect-the-dot) figures, which are spatial networks drawn over the fixed background of stars. We analyse 1802 line figures from 56 sky cultures spanning all continents, in terms of their network, spatial, and brightness features, and ask what associations exist between these visual features and culture type or sky region. First, an embedded map of constellations is learnt, to show clusters of line figures. We then form the network of constellations (as linked by their similarity), to study how similar cultures are by computing their assortativity (or homophily) over the network. Finally, we measure the diversity (or entropy) index for the set of constellations drawn per sky region. Our results show distinct types of line figures, and that many folk astronomies with oral traditions have widespread similarities in constellation design, which do not align with cultural ancestry. In a minority of sky regions, certain line designs appear universal, but this is not the norm: in the majority of sky regions, the line geometries are diverse.

Most impressively, Samasource has overcome a problem that most Silicon Valley firms are famously grappling with. Just over half of their workforce is made up of women, a remarkable feat in a country where starting a family more often than not rules out a career for the mother. Here, a lactation room, up to 90 days maternity leave, and flexibility around shift patterns makes the firm a stand-out example of inclusivity not just in Kenya, but globally.

Can a machine be an inventor? After the courts said no, a computer scientist is once more trying to have an artificial intelligence considered an inventor in the eyes of the law. In August, the U.S. Federal Circuit Court of Appeals issued a decision that AI cannot be listed as the inventor on a patent registration. The case before the court--Thaler v. Vidal--was either a gimmick that could be dismissed with a simple reading of U.S. patent law or one that strikes at the heart of a metaphysical question with crucial implications for the future of innovation. In Thaler v. Vidal, Stephen Thaler challenged the refusal of the U.S. Patent and Trademark Office to issue a patent registration for an invention Thaler claims was created by an artificial intelligence device called Device for Autonomous Bootstrapping of Unified Sentience, or DABUS.

We introduce AARGH, an end-to-end task-oriented dialog system combining retrieval and generative approaches in a single model, aiming at improving dialog management and lexical diversity of outputs. The model features a new response selection method based on an action-aware training objective and a simplified single-encoder retrieval architecture which allow us to build an end-to-end retrieval-enhanced generation model where retrieval and generation share most of the parameters. On the MultiWOZ dataset, we show that our approach produces more diverse outputs while maintaining or improving state tracking and context-to-response generation performance, compared to state-of-the-art baselines.

It is well recognized that sensory perceptions and language have interconnections through numerous studies in psychology, neuroscience, and sensorial linguistics. Set in this rich context we ask whether the use of sensorial language in writings is part of linguistic style? This question is important from the view of stylometrics research where a rich set of language features have been explored, but with insufficient attention given to features related to sensorial language. Taking this as the goal we explore several angles about sensorial language and style in collections of lyrics, novels, and poetry. We find, for example, that individual use of sensorial language is not a random phenomenon; choice is likely involved. Also, sensorial style is generally stable over time - the shifts are extremely small. Moreover, style can be extracted from just a few hundred sentences that have sensorial terms. We also identify representative and distinctive features within each genre. For example, we observe that 4 of the top 6 representative features in novels collection involved individuals using olfactory language where we expected them to use non-olfactory language.

The availability of well-curated datasets has driven the success of Machine Learning (ML) models. Despite the increased access to earth observation data for agriculture, there is a scarcity of curated, labelled datasets, which limits the potential of its use in training ML models for remote sensing (RS) in agriculture. To this end, we introduce a first-of-its-kind dataset, SICKLE, having time-series images at different spatial resolutions from 3 different satellites, annotated with multiple key cropping parameters for paddy cultivation for the Cauvery Delta region in Tamil Nadu, India. The dataset comprises of 2,398 season-wise samples from 388 unique plots distributed across 4 districts of the Delta. The dataset covers multi-spectral, thermal and microwave data between the time period January 2018-March 2021. The paddy samples are annotated with 4 key cropping parameters, i.e. sowing date, transplanting date, harvesting date and crop yield. This is one of the first studies to consider the growing season (using sowing and harvesting dates) as part of a dataset. We also propose a yield prediction strategy that uses time-series data generated based on the observed growing season and the standard seasonal information obtained from Tamil Nadu Agricultural University for the region. The consequent performance improvement highlights the impact of ML techniques that leverage domain knowledge that are consistent with standard practices followed by farmers in a specific region. We benchmark the dataset on 3 separate tasks, namely crop type, phenology date (sowing, transplanting, harvesting) and yield prediction, and develop an end-to-end framework for predicting key crop parameters in a real-world setting.

Traditional federated optimization methods perform poorly with heterogeneous data (ie, accuracy reduction), especially for highly skewed data. In this paper, we investigate the label distribution skew in FL, where the distribution of labels varies across clients. First, we investigate the label distribution skew from a statistical view. We demonstrate both theoretically and empirically that previous methods based on softmax cross-entropy are not suitable, which can result in local models heavily overfitting to minority classes and missing classes. Additionally, we theoretically introduce a deviation bound to measure the deviation of the gradient after local update. At last, we propose FedLC (\textbf {Fed} erated learning via\textbf {L} ogits\textbf {C} alibration), which calibrates the logits before softmax cross-entropy according to the probability of occurrence of each class. FedLC applies a fine-grained calibrated cross-entropy loss to local update by adding a pairwise label margin. Extensive experiments on federated datasets and real-world datasets demonstrate that FedLC leads to a more accurate global model and much improved performance. Furthermore, integrating other FL methods into our approach can further enhance the performance of the global model.

Annotated data is an essential ingredient in natural language processing for training and evaluating machine learning models. It is therefore very desirable for the annotations to be of high quality. Recent work, however, has shown that several popular datasets contain a surprising amount of annotation errors or inconsistencies. To alleviate this issue, many methods for annotation error detection have been devised over the years. While researchers show that their approaches work well on their newly introduced datasets, they rarely compare their methods to previous work or on the same datasets. This raises strong concerns on methods' general performance and makes it difficult to asses their strengths and weaknesses. We therefore reimplement 18 methods for detecting potential annotation errors and evaluate them on 9 English datasets for text classification as well as token and span labeling. In addition, we define a uniform evaluation setup including a new formalization of the annotation error detection task, evaluation protocol and general best practices. To facilitate future research and reproducibility, we release our datasets and implementations in an easy-to-use and open source software package.