Journal articles on the topic 'NLI'
To see the other types of publications on this topic, follow the link: NLI.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 journal articles for your research on the topic 'NLI.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse journal articles on a wide variety of disciplines and organise your bibliography correctly.
1
Chen, Weixing. "Polynomial rings over NLI rings need not be NLI." Studia Scientiarum Mathematicarum Hungarica 52, no. 1 (March 1, 2015): 129–33. http://dx.doi.org/10.1556/sscmath.52.2015.1.1305.
Full textAbstract:
It is proved that there exists an NI ring R over which the polynomial ring R[x] is not an NLI ring. This answers an open question of Qu and Wei (Stud. Sci. Math. Hung., 51(2), 2014) in the negative. Moreover a sufficient condition of R[x] to be an NLI ring is included for an NLI ring R.
2
Jurata, L. W., and G. N. Gill. "Functional analysis of the nuclear LIM domain interactor NLI." Molecular and Cellular Biology 17, no. 10 (October 1997): 5688–98. http://dx.doi.org/10.1128/mcb.17.10.5688.
Full textAbstract:
LIM homeodomain and LIM-only (LMO) transcription factors contain two tandemly arranged Zn2+-binding LIM domains capable of mediating protein-protein interactions. These factors have restricted patterns of expression, are found in invertebrates as well as vertebrates, and are required for cell type specification in a variety of developing tissues. A recently identified, widely expressed protein, NLI, binds with high affinity to the LIM domains of LIM homeodomain and LMO proteins in vitro and in vivo. In this study, a 38-amino-acid fragment of NLI was found to be sufficient for the association of NLI with nuclear LIM domains. In addition, NLI was shown to form high affinity homodimers through the amino-terminal 200 amino acids, but dimerization of NLI was not required for association with the LIM homeodomain protein Lmxl. Chemical cross-linking analysis revealed higher-order complexes containing multiple NLI molecules bound to Lmx1, indicating that dimerization of NLI does not interfere with LIM domain interactions. Additionally, NLI formed complexes with Lmx1 on the rat insulin I promoter and inhibited the LIM domain-dependent synergistic transcriptional activation by Lmx1 and the basic helix-loop-helix protein E47 from the rat insulin I minienhancer. These studies indicate that NLI contains at least two functionally independent domains and may serve as a negative regulator of synergistic transcriptional responses which require direct interaction via LIM domains. Thus, NLI may regulate the transcriptional activity of LIM homeodomain proteins by determining specific partner interactions.
3
Renjit, Sara, and Sumam Idicula. "Natural language inference for Malayalam language using language agnostic sentence representation." PeerJ Computer Science 7 (May 4, 2021): e508. http://dx.doi.org/10.7717/peerj-cs.508.
Full textAbstract:
Natural language inference (NLI) is an essential subtask in many natural language processing applications. It is a directional relationship from premise to hypothesis. A pair of texts is defined as entailed if a text infers its meaning from the other text. The NLI is also known as textual entailment recognition, and it recognizes entailed and contradictory sentences in various NLP systems like Question Answering, Summarization and Information retrieval systems. This paper describes the NLI problem attempted for a low resource Indian language Malayalam, the regional language of Kerala. More than 30 million people speak this language. The paper is about the Malayalam NLI dataset, named MaNLI dataset, and its application of NLI in Malayalam language using different models, namely Doc2Vec (paragraph vector), fastText, BERT (Bidirectional Encoder Representation from Transformers), and LASER (Language Agnostic Sentence Representation). Our work attempts NLI in two ways, as binary classification and as multiclass classification. For both the classifications, LASER outperformed the other techniques. For multiclass classification, NLI using LASER based sentence embedding technique outperformed the other techniques by a significant margin of 12% accuracy. There was also an accuracy improvement of 9% for LASER based NLI system for binary classification over the other techniques.
4
Wang, Haohan, Da Sun, and Eric P. Xing. "What if We Simply Swap the Two Text Fragments? A Straightforward yet Effective Way to Test the Robustness of Methods to Confounding Signals in Nature Language Inference Tasks." Proceedings of the AAAI Conference on Artificial Intelligence 33 (July 17, 2019): 7136–43. http://dx.doi.org/10.1609/aaai.v33i01.33017136.
Full textAbstract:
Nature language inference (NLI) task is a predictive task of determining the inference relationship of a pair of natural language sentences. With the increasing popularity of NLI, many state-of-the-art predictive models have been proposed with impressive performances. However, several works have noticed the statistical irregularities in the collected NLI data set that may result in an over-estimated performance of these models and proposed remedies. In this paper, we further investigate the statistical irregularities, what we refer as confounding factors, of the NLI data sets. With the belief that some NLI labels should preserve under swapping operations, we propose a simple yet effective way (swapping the two text fragments) of evaluating the NLI predictive models that naturally mitigate the observed problems. Further, we continue to train the predictive models with our swapping manner and propose to use the deviation of the model’s evaluation performances under different percentages of training text fragments to be swapped to describe the robustness of a predictive model. Our evaluation metrics leads to some interesting understandings of recent published NLI methods. Finally, we also apply the swapping operation on NLI models to see the effectiveness of this straightforward method in mitigating the confounding factor problems in training generic sentence embeddings for other NLP transfer tasks.
5
Qu, Yinchun, and Junchao Wei. "Rings whose nilpotent elements form a Lie ideal." Studia Scientiarum Mathematicarum Hungarica 51, no. 2 (June 1, 2014): 271–84. http://dx.doi.org/10.1556/sscmath.51.2014.2.1279.
Full textAbstract:
A ring R is called NLI (rings whose nilpotent elements form a Lie ideal) if for each a ∈ N(R) and b ∈ R, ab − ba ∈ N(R). Clearly, NI rings are NLI. In this note, many properties of NLI rings are studied. The main results we obtain are the following: (1) NLI rings are directly finite and left min-abel; (2) If R is a NLI ring, then (a) R is a strongly regular ring if and only if R is a Von Neumann regular ring; (b) R is (weakly) exchange if and only if R is (weakly) clean; (c) R is a reduced ring if and only if R is a n-regular ring; (3) If R is a NLI left MC2 ring whose singular simple left modules are Wnil-injective, then R is reduced.
6
Wang, Xiaoyan, Pavan Kapanipathi, Ryan Musa, Mo Yu, Kartik Talamadupula, Ibrahim Abdelaziz, Maria Chang, et al. "Improving Natural Language Inference Using External Knowledge in the Science Questions Domain." Proceedings of the AAAI Conference on Artificial Intelligence 33 (July 17, 2019): 7208–15. http://dx.doi.org/10.1609/aaai.v33i01.33017208.
Full textAbstract:
Natural Language Inference (NLI) is fundamental to many Natural Language Processing (NLP) applications including semantic search and question answering. The NLI problem has gained significant attention due to the release of large scale, challenging datasets. Present approaches to the problem largely focus on learning-based methods that use only textual information in order to classify whether a given premise entails, contradicts, or is neutral with respect to a given hypothesis. Surprisingly, the use of methods based on structured knowledge – a central topic in artificial intelligence – has not received much attention vis-a-vis the NLI problem. While there are many open knowledge bases that contain various types of reasoning information, their use for NLI has not been well explored. To address this, we present a combination of techniques that harness external knowledge to improve performance on the NLI problem in the science questions domain. We present the results of applying our techniques on text, graph, and text-and-graph based models; and discuss the implications of using external knowledge to solve the NLI problem. Our model achieves close to state-of-the-art performance for NLI on the SciTail science questions dataset.
7
Song, Meina, Wen Zhao, and E. HaiHong. "KGAnet: a knowledge graph attention network for enhancing natural language inference." Neural Computing and Applications 32, no. 18 (June 26, 2020): 14963–73. http://dx.doi.org/10.1007/s00521-020-04851-5.
Full textAbstract:
Abstract Natural language inference (NLI) is the basic task of many applications such as question answering and paraphrase recognition. Existing methods have solved the key issue of how the NLI model can benefit from external knowledge. Inspired by this, we attempt to further explore the following two problems: (1) how to make better use of external knowledge when the total amount of such knowledge is constant and (2) how to bring external knowledge to the NLI model more conveniently in the application scenario. In this paper, we propose a novel joint training framework that consists of a modified graph attention network, called the knowledge graph attention network, and an NLI model. We demonstrate that the proposed method outperforms the existing method which introduces external knowledge, and we improve the performance of multiple NLI models without additional external knowledge.
8
Song, Haoyu, Wei-Nan Zhang, Jingwen Hu, and Ting Liu. "Generating Persona Consistent Dialogues by Exploiting Natural Language Inference." Proceedings of the AAAI Conference on Artificial Intelligence 34, no. 05 (April 3, 2020): 8878–85. http://dx.doi.org/10.1609/aaai.v34i05.6417.
Full textAbstract:
Consistency is one of the major challenges faced by dialogue agents. A human-like dialogue agent should not only respond naturally, but also maintain a consistent persona. In this paper, we exploit the advantages of natural language inference (NLI) technique to address the issue of generating persona consistent dialogues. Different from existing work that re-ranks the retrieved responses through an NLI model, we cast the task as a reinforcement learning problem and propose to exploit the NLI signals from response-persona pairs as rewards for the process of dialogue generation. Specifically, our generator employs an attention-based encoder-decoder to generate persona-based responses. Our evaluator consists of two components: an adversarially trained naturalness module and an NLI based consistency module. Moreover, we use another well-performed NLI model in the evaluation of persona-consistency. Experimental results on both human and automatic metrics, including the model-based consistency evaluation, demonstrate that the proposed approach outperforms strong generative baselines, especially in the persona-consistency of generated responses.
9
Antoine, R., Ph Dugourd, D. Rayane, and M. Broyer. "Dissociation pathways and binding energies of (LiH)nLi+ and (LiH)nLi+3 clusters." Journal of Chemical Physics 104, no. 1 (January 1996): 110–19. http://dx.doi.org/10.1063/1.470880.
Full text
10
Zhou, Kang, Qiao Qiao, Yuepei Li, and Qi Li. "Improving Distantly Supervised Relation Extraction by Natural Language Inference." Proceedings of the AAAI Conference on Artificial Intelligence 37, no. 11 (June 26, 2023): 14047–55. http://dx.doi.org/10.1609/aaai.v37i11.26644.
Full textAbstract:
To reduce human annotations for relation extraction (RE) tasks, distantly supervised approaches have been proposed, while struggling with low performance. In this work, we propose a novel DSRE-NLI framework, which considers both distant supervision from existing knowledge bases and indirect supervision from pretrained language models for other tasks. DSRE-NLI energizes an off-the-shelf natural language inference (NLI) engine with a semi-automatic relation verbalization (SARV) mechanism to provide indirect supervision and further consolidates the distant annotations to benefit multi-classification RE models. The NLI-based indirect supervision acquires only one relation verbalization template from humans as a semantically general template for each relationship, and then the template set is enriched by high-quality textual patterns automatically mined from the distantly annotated corpus. With two simple and effective data consolidation strategies, the quality of training data is substantially improved. Extensive experiments demonstrate that the proposed framework significantly improves the SOTA performance (up to 7.73% of F1) on distantly supervised RE benchmark datasets. Our code is available at https://github.com/kangISU/DSRE-NLI.
11
Chen, Yanran, and Steffen Eger. "MENLI: Robust Evaluation Metrics from Natural Language Inference." Transactions of the Association for Computational Linguistics 11 (2023): 804–25. http://dx.doi.org/10.1162/tacl_a_00576.
Full textAbstract:
Abstract Recently proposed BERT-based evaluation metrics for text generation perform well on standard benchmarks but are vulnerable to adversarial attacks, e.g., relating to information correctness. We argue that this stems (in part) from the fact that they are models of semantic similarity. In contrast, we develop evaluation metrics based on Natural Language Inference (NLI), which we deem a more appropriate modeling. We design a preference-based adversarial attack framework and show that our NLI based metrics are much more robust to the attacks than the recent BERT-based metrics. On standard benchmarks, our NLI based metrics outperform existing summarization metrics, but perform below SOTA MT metrics. However, when combining existing metrics with our NLI metrics, we obtain both higher adversarial robustness (15%–30%) and higher quality metrics as measured on standard benchmarks (+5% to 30%).
12
Laban, Philippe, Tobias Schnabel, Paul N. Bennett, and Marti A. Hearst. "SummaC: Re-Visiting NLI-based Models for Inconsistency Detection in Summarization." Transactions of the Association for Computational Linguistics 10 (2022): 163–77. http://dx.doi.org/10.1162/tacl_a_00453.
Full textAbstract:
Abstract In the summarization domain, a key requirement for summaries is to be factually consistent with the input document. Previous work has found that natural language inference (NLI) models do not perform competitively when applied to inconsistency detection. In this work, we revisit the use of NLI for inconsistency detection, finding that past work suffered from a mismatch in input granularity between NLI datasets (sentence-level), and inconsistency detection (document level). We provide a highly effective and light-weight method called SummaCConv that enables NLI models to be successfully used for this task by segmenting documents into sentence units and aggregating scores between pairs of sentences. We furthermore introduce a new benchmark called SummaC (Summary Consistency) which consists of six large inconsistency detection datasets. On this dataset, SummaCConv obtains state-of-the-art results with a balanced accuracy of 74.4%, a 5% improvement compared with prior work.
13
Percha, Bethany, Kereeti Pisapati, Cynthia Gao, and Hank Schmidt. "Natural language inference for curation of structured clinical registries from unstructured text." Journal of the American Medical Informatics Association 29, no. 1 (November 13, 2021): 97–108. http://dx.doi.org/10.1093/jamia/ocab243.
Full textAbstract:
Abstract Objective Clinical registries—structured databases of demographic, diagnosis, and treatment information—play vital roles in retrospective studies, operational planning, and assessment of patient eligibility for research, including clinical trials. Registry curation, a manual and time-intensive process, is always costly and often impossible for rare or underfunded diseases. Our goal was to evaluate the feasibility of natural language inference (NLI) as a scalable solution for registry curation. Materials and Methods We applied five state-of-the-art, pretrained, deep learning-based NLI models to clinical, laboratory, and pathology notes to infer information about 43 different breast oncology registry fields. Model inferences were evaluated against a manually curated, 7439 patient breast oncology research database. Results NLI models showed considerable variation in performance, both within and across fields. One model, ALBERT, outperformed the others (BART, RoBERTa, XLNet, and ELECTRA) on 22 out of 43 fields. A detailed error analysis revealed that incorrect inferences primarily arose through models' tendency to misinterpret historical findings, as well as confusion based on abbreviations and subtle term variants common in clinical text. Discussion and Conclusion Traditional natural language processing methods require specially annotated training sets or the construction of a separate model for each registry field. In contrast, a single pretrained NLI model can curate dozens of different fields simultaneously. Surprisingly, NLI methods remain unexplored in the clinical domain outside the realm of shared tasks and benchmarks. Modern NLI models could increase the efficiency of registry curation, even when applied “out of the box” with no additional training.
14
Kiefer, Christine M., Jongjoo Lee, Chunhui Hou, Ryan K. Dale, Y. Terry Lee, Emily R. Meier, Jeffrey L. Miller, and Ann Dean. "Distinct Ldb1/NLI complexes orchestrate γ-globin repression and reactivation through ETO2 in human adult erythroid cells." Blood 118, no. 23 (December 1, 2011): 6200–6208. http://dx.doi.org/10.1182/blood-2011-06-363101.
Full textAbstract:
Abstract The Ldb1/GATA-1/TAL1/LMO2 complex mediates long-range interaction between the β-globin locus control region (LCR) and gene in adult mouse erythroid cells, but whether this complex mediates chromatin interactions at other developmental stages or in human cells is unknown. We investigated NLI (Ldb1 homolog) complex occupancy and chromatin conformation of the β-globin locus in human erythroid cells. In addition to the LCR, we found robust NLI complex occupancy at a site downstream of the Aγ-globin gene within sequences of BGL3, an intergenic RNA transcript. In cells primarily transcribing β-globin, BGL3 is not transcribed and BGL3 sequences are occupied by NLI core complex members, together with corepressor ETO2 and by γ-globin repressor BCL11A. The LCR and β-globin gene establish proximity in these cells. In contrast, when γ-globin transcription is reactivated in these cells, ETO2 participation in the NLI complex at BGL3 is diminished, as is BCL11A occupancy, and both BGL3 and γ-globin are transcribed. In these cells, proximity between the BGL3/γ-globin region and the LCR is established. We conclude that alternative NLI complexes mediate γ-globin transcription or silencing through long-range LCR interactions involving an intergenic site of noncoding RNA transcription and that ETO2 is critical to this process.
15
Ni, Pin, Yuming Li, Gangmin Li, and Victor Chang. "A Hybrid Siamese Neural Network for Natural Language Inference in Cyber-Physical Systems." ACM Transactions on Internet Technology 21, no. 2 (March 10, 2021): 1–25. http://dx.doi.org/10.1145/3418208.
Full textAbstract:
Cyber-Physical Systems (CPS), as a multi-dimensional complex system that connects the physical world and the cyber world, has a strong demand for processing large amounts of heterogeneous data. These tasks also include Natural Language Inference (NLI) tasks based on text from different sources. However, the current research on natural language processing in CPS does not involve exploration in this field. Therefore, this study proposes a Siamese Network structure that combines Stacked Residual Long Short-Term Memory (bidirectional) with the Attention mechanism and Capsule Network for the NLI module in CPS, which is used to infer the relationship between text/language data from different sources. This model is mainly used to implement NLI tasks and conduct a detailed evaluation in three main NLI benchmarks as the basic semantic understanding module in CPS. Comparative experiments prove that the proposed method achieves competitive performance, has a certain generalization ability, and can balance the performance and the number of trained parameters.
16
Mitra, Arindam, Ishan Shrivastava, and Chitta Baral. "Enhancing Natural Language Inference Using New and Expanded Training Data Sets and New Learning Models." Proceedings of the AAAI Conference on Artificial Intelligence 34, no. 05 (April 3, 2020): 8504–11. http://dx.doi.org/10.1609/aaai.v34i05.6371.
Full textAbstract:
Natural Language Inference (NLI) plays an important role in many natural language processing tasks such as question answering. However, existing NLI modules that are trained on existing NLI datasets have several drawbacks. For example, they do not capture the notion of entity and role well and often end up making mistakes such as “Peter signed a deal” can be inferred from “John signed a deal”. As part of this work, we have developed two datasets that help mitigate such issues and make the systems better at understanding the notion of “entities” and “roles”. After training the existing models on the new dataset we observe that the existing models do not perform well on one of the new benchmark. We then propose a modification to the “word-to-word” attention function which has been uniformly reused across several popular NLI architectures. The resulting models perform as well as their unmodified counterparts on the existing benchmarks and perform significantly well on the new benchmarks that emphasize “roles” and “entities”.
17
Liu, Hanmeng, Leyang Cui, Jian Liu, and Yue Zhang. "Natural Language Inference in Context - Investigating Contextual Reasoning over Long Texts." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 15 (May 18, 2021): 13388–96. http://dx.doi.org/10.1609/aaai.v35i15.17580.
Full textAbstract:
Natural language inference (NLI) is a fundamental NLP task, investigating the entailment relationship between two texts. Popular NLI datasets present the task at sentence-level. While adequate for testing semantic representations, they fall short for testing contextual reasoning over long texts, which is a natural part of the human inference process. We introduce ConTRoL, a new dataset for ConTextual Reasoning over Long texts. Consisting of 8,325 expert-designed "context-hypothesis" pairs with gold labels, ConTRoL is a passage-level NLI dataset with a focus on complex contextual reasoning types such as logical reasoning. It is derived from competitive selection and recruitment test (verbal reasoning test) for police recruitment, with expert level quality. Compared with previous NLI benchmarks, the materials in ConTRoL are much more challenging, involving a range of reasoning types. Empirical results show that state-of-the-art language models perform by far worse than educated humans. Our dataset can also serve as a testing-set for downstream tasks like checking the factual correctness of summaries.
18
Eleftheriadis, Petros, Isidoros Perikos, and Ioannis Hatzilygeroudis. "Evaluating Deep Learning Techniques for Natural Language Inference." Applied Sciences 13, no. 4 (February 16, 2023): 2577. http://dx.doi.org/10.3390/app13042577.
Full textAbstract:
Natural language inference (NLI) is one of the most important natural language understanding (NLU) tasks. NLI expresses the ability to infer information during spoken or written communication. The NLI task concerns the determination of the entailment relation of a pair of sentences, called the premise and hypothesis. If the premise entails the hypothesis, the pair is labeled as an “entailment”. If the hypothesis contradicts the premise, the pair is labeled a “contradiction”, and if there is not enough information to infer a relationship, the pair is labeled as “neutral”. In this paper, we present experimentation results of using modern deep learning (DL) models, such as the pre-trained transformer BERT, as well as additional models that relay on LSTM networks, for the NLI task. We compare five DL models (and variations of them) on eight widely used NLI datasets. We trained and fine-tuned the hyperparameters for each model to achieve the best performance for each dataset, where we achieved some state-of-the-art results. Next, we examined the inference ability of the models on the BreakingNLI dataset, which evaluates the model’s ability to recognize lexical inferences. Finally, we tested the generalization power of our models across all the NLI datasets. The results of the study are quite interesting. In the first part of our experimentation, the results indicate the performance advantage of the pre-trained transformers BERT, RoBERTa, and ALBERT over other deep learning models. This became more evident when they were tested on the BreakingNLI dataset. We also see a pattern of improved performance when the larger models are used. However, ALBERT, given that it has 18 times fewer parameters, achieved quite remarkable performance.
19
Anatskaya, Olga V., Andrey L. Runov, Sergey V. Ponomartsev, Maxim S. Vonsky, Artem U. Elmuratov, and Alexander E. Vinogradov. "Long-Term Transcriptomic Changes and Cardiomyocyte Hyperpolyploidy after Lactose Intolerance in Neonatal Rats." International Journal of Molecular Sciences 24, no. 8 (April 11, 2023): 7063. http://dx.doi.org/10.3390/ijms24087063.
Full textAbstract:
Many cardiovascular diseases originate from growth retardation, inflammation, and malnutrition during early postnatal development. The nature of this phenomenon is not completely understood. Here we aimed to verify the hypothesis that systemic inflammation triggered by neonatal lactose intolerance (NLI) may exert long-term pathologic effects on cardiac developmental programs and cardiomyocyte transcriptome regulation. Using the rat model of NLI triggered by lactase overloading with lactose and the methods of cytophotometry, image analysis, and mRNA-seq, we evaluated cardiomyocyte ploidy, signs of DNA damage, and NLI-associated long-term transcriptomic changes of genes and gene modules that differed qualitatively (i.e., were switched on or switched off) in the experiment vs. the control. Our data indicated that NLI triggers the long-term animal growth retardation, cardiomyocyte hyperpolyploidy, and extensive transcriptomic rearrangements. Many of these rearrangements are known as manifestations of heart pathologies, including DNA and telomere instability, inflammation, fibrosis, and reactivation of fetal gene program. Moreover, bioinformatic analysis identified possible causes of these pathologic traits, including the impaired signaling via thyroid hormone, calcium, and glutathione. We also found transcriptomic manifestations of increased cardiomyocyte polyploidy, such as the induction of gene modules related to open chromatin, e.g., “negative regulation of chromosome organization”, “transcription” and “ribosome biogenesis”. These findings suggest that ploidy-related epigenetic alterations acquired in the neonatal period permanently rewire gene regulatory networks and alter cardiomyocyte transcriptome. Here we provided first evidence indicating that NLI can be an important trigger of developmental programming of adult cardiovascular disease. The obtained results can help to develop preventive strategies for reducing the NLI-associated adverse effects of inflammation on the developing cardiovascular system.
20
Malmasi, Shervin, and Mark Dras. "Native Language Identification With Classifier Stacking and Ensembles." Computational Linguistics 44, no. 3 (September 2018): 403–46. http://dx.doi.org/10.1162/coli_a_00323.
Full textAbstract:
Ensemble methods using multiple classifiers have proven to be among the most successful approaches for the task of Native Language Identification (NLI), achieving the current state of the art. However, a systematic examination of ensemble methods for NLI has yet to be conducted. Additionally, deeper ensemble architectures such as classifier stacking have not been closely evaluated. We present a set of experiments using three ensemble-based models, testing each with multiple configurations and algorithms. This includes a rigorous application of meta-classification models for NLI, achieving state-of-the-art results on several large data sets, evaluated in both intra-corpus and cross-corpus modes.
21
Richardson, Kyle, Hai Hu, Lawrence Moss, and Ashish Sabharwal. "Probing Natural Language Inference Models through Semantic Fragments." Proceedings of the AAAI Conference on Artificial Intelligence 34, no. 05 (April 3, 2020): 8713–21. http://dx.doi.org/10.1609/aaai.v34i05.6397.
Full textAbstract:
Do state-of-the-art models for language understanding already have, or can they easily learn, abilities such as boolean coordination, quantification, conditionals, comparatives, and monotonicity reasoning (i.e., reasoning about word substitutions in sentential contexts)? While such phenomena are involved in natural language inference (NLI) and go beyond basic linguistic understanding, it is unclear the extent to which they are captured in existing NLI benchmarks and effectively learned by models. To investigate this, we propose the use of semantic fragments—systematically generated datasets that each target a different semantic phenomenon—for probing, and efficiently improving, such capabilities of linguistic models. This approach to creating challenge datasets allows direct control over the semantic diversity and complexity of the targeted linguistic phenomena, and results in a more precise characterization of a model's linguistic behavior. Our experiments, using a library of 8 such semantic fragments, reveal two remarkable findings: (a) State-of-the-art models, including BERT, that are pre-trained on existing NLI benchmark datasets perform poorly on these new fragments, even though the phenomena probed here are central to the NLI task; (b) On the other hand, with only a few minutes of additional fine-tuning—with a carefully selected learning rate and a novel variation of “inoculation”—a BERT-based model can master all of these logic and monotonicity fragments while retaining its performance on established NLI benchmarks.
22
Yanaka, Hitomi, and Koji Mineshima. "Compositional Evaluation on Japanese Textual Entailment and Similarity." Transactions of the Association for Computational Linguistics 10 (2022): 1266–84. http://dx.doi.org/10.1162/tacl_a_00518.
Full textAbstract:
Abstract Natural Language Inference (NLI) and Semantic Textual Similarity (STS) are widely used benchmark tasks for compositional evaluation of pre-trained language models. Despite growing interest in linguistic universals, most NLI/STS studies have focused almost exclusively on English. In particular, there are no available multilingual NLI/STS datasets in Japanese, which is typologically different from English and can shed light on the currently controversial behavior of language models in matters such as sensitivity to word order and case particles. Against this background, we introduce JSICK, a Japanese NLI/STS dataset that was manually translated from the English dataset SICK. We also present a stress-test dataset for compositional inference, created by transforming syntactic structures of sentences in JSICK to investigate whether language models are sensitive to word order and case particles. We conduct baseline experiments on different pre-trained language models and compare the performance of multilingual models when applied to Japanese and other languages. The results of the stress-test experiments suggest that the current pre-trained language models are insensitive to word order and case marking.
23
Reddy, Vathala Nagi, Tanikonda Prasanth, Nakka Venkata Karthik, Puram Lohith Kumar, Yekollu Chakradhar, K. V. Narayana, K. Sowjan kumar, Dr Y. Prakash, and Dr K. Venkateswarlu. "A Novel Multilevel Converter with Stable Voltage to The Renewable Energy Systems." International Journal of Innovative Research in Engineering and Management 8, no. 6 (November 30, 2021): 1135–39. http://dx.doi.org/10.55524/ijirem.2021.8.6.224.
Full textAbstract:
In this proposed paper, a novel nine-level inverter (NLI) topology is proposed, which uses fewer components to provide nine voltage levels in the output. Ten semiconductor switches, four capacitors, and one dc voltage source are included in this NLI. Through various logical arrangements of the dc source and capacitors and the use of power semiconductor switches, the nine voltage levels are generated. In this case, only two switches can conduct simultaneously in any mode. As a result, the system's efficiency is increased and conduction losses are decreased. The low total standing voltage (TSV) of the power semiconductor switches is another crucial aspect of this design. Gate signals are produced using a suggested modulation approach to lessen the power switches' switching loss. Several Simulink simulations are run in order to validate the NLI. The suggested inverter has a much lower requirement for components and dc sources as compared to conventional topologies. In order to increase the efficiency of the power generating process in renewable energy systems, the proposed NLI structure can be implemented.
24
Nie, Yixin, Yicheng Wang, and Mohit Bansal. "Analyzing Compositionality-Sensitivity of NLI Models." Proceedings of the AAAI Conference on Artificial Intelligence 33 (July 17, 2019): 6867–74. http://dx.doi.org/10.1609/aaai.v33i01.33016867.
Full textAbstract:
Success in natural language inference (NLI) should require a model to understand both lexical and compositional semantics. However, through adversarial evaluation, we find that several state-of-the-art models with diverse architectures are over-relying on the former and fail to use the latter. Further, this compositionality unawareness is not reflected via standard evaluation on current datasets. We show that removing RNNs in existing models or shuffling input words during training does not induce large performance loss despite the explicit removal of compositional information. Therefore, we propose a compositionality-sensitivity testing setup that analyzes models on natural examples from existing datasets that cannot be solved via lexical features alone (i.e., on which a bag-of-words model gives a high probability to one wrong label), hence revealing the models’ actual compositionality awareness. We show that this setup not only highlights the limited compositional ability of current NLI models, but also differentiates model performance based on design, e.g., separating shallow bag-of-words models from deeper, linguistically-grounded tree-based models. Our evaluation setup is an important analysis tool: complementing currently existing adversarial and linguistically driven diagnostic evaluations, and exposing opportunities for future work on evaluating models’ compositional understanding.
25
Liang, Ze, Yueyao Wang, Fuyue Sun, Hong Jiang, Jiao Huang, Jiashu Shen, Feili Wei, and Shuangcheng Li. "Exploring the Combined Effect of Urbanization and Climate Variability on Urban Vegetation: A Multi-Perspective Study Based on More than 3000 Cities in China." Remote Sensing 12, no. 8 (April 22, 2020): 1328. http://dx.doi.org/10.3390/rs12081328.
Full textAbstract:
More than 3000 cities in China were used to study the effect of urbanization and local climate variability on urban vegetation across different geographical and urbanization conditions. The national scale estimation shows that China’s urban vegetation depicts a trend of degradation from 2000 to 2015, especially in developed areas such as the Yangtze River Delta. According to the panel models, the increase of precipitation (PREC), solar radiation (SRAD), air temperature (TEMP), and specific humidity (SHUM) all enhance urban vegetation, while nighttime light intensity (NLI), population density (POPDEN), and fractal dimension (FRAC) do the opposite. The effects change along the East–West gradient; the influences of PREC and SHUM become greater, while those of TEMP, SRAD, NLI, AREA, and FRAC become smaller. PREC, SHUM, and SRAD play the most important roles in Northeast, Central, and North China, respectively. The role of FRAC and NLI in East China is much greater than in other regions. POPDEN remains influential across all altitudes, while FRAC affects only low-altitude cities. NLI plays a greater role in larger cities, while FRAC and POPDEN are the opposite. In cities outside of the five major urban agglomerations, PREC has a great influence while the key factors are more diversified inside.
26
Guo, Zhicheng, Tianxin Hao, and Xinlinzi Wang. "Enhancing Children's Academic Learning Experience: Natural Language Interface Designs in Educational Application." Highlights in Science, Engineering and Technology 85 (March 13, 2024): 621–26. http://dx.doi.org/10.54097/jd0vp179.
Full textAbstract:
User Interaction (UI) designs, a crucial part of the modern internet industry, have been envolving and becoming more and more sophisticated and advanced. Recently, with the development of Machine Learning in Computer Science, Natural Language Interface (NLI) designs, a new division of the UI designs, appears into public’s view. With its special focus on using Natural Language Processing (NLP) to enable applications respond and execute useful actions in response to users’ input, UI designers are able to implement an array of new functions into the application, allowing children (12 to 18 years old) to better interact with the applications. This research primarily focuses on how NLI designs in educational settings are able to enhance their learning experience and academic performance. In particular, with a deep literary analysis, this research incorporates several important findings about NLI designs’ positive effects on children’s academic learning experience and performance. Through thorough investigations, this research concludes that while there are several technical limitations on NLI designs due to the limited level of technology, they can still facilitate children’s learning experience and academic performance by satisfying children, letting them stay longer in the application, and providing them a better guide, which enables them to have a better navigation in the application’s learning environment.
27
Kyle, Kristopher, Scott A. Crossley, and YouJin Kim. "Native language identification and writing proficiency." International Journal of Learner Corpus Research 1, no. 2 (September 14, 2015): 187–209. http://dx.doi.org/10.1075/ijlcr.1.2.01kyl.
Full textAbstract:
This study evaluates the impact of writing proficiency on native language identification (NLI), a topic that has important implications for the generalizability of NLI models and detection-based arguments for cross-linguistic influence (Jarvis 2010, 2012; CLI). The study uses multinomial logistic regression to classify the first language (L1) group membership of essays at two proficiency levels based on systematic lexical and phrasal choices made by members of five L1 groups. The results indicate that lower proficiency essays are significantly easier to classify than higher proficiency essays, suggesting that lower proficiency writers make lexical and phrasal choices that are more similar to other lower proficiency writers that share an L1 than higher proficiency writers that share an L1. A close analysis of the findings also indicates that the relationship between NLI accuracy and proficiency differed across L1 groups.
28
Rice, Mabel L., Catherine L. Taylor, Stephen R. Zubrick, Lesa Hoffman, and Kathleen K. Earnest. "Heritability of Specific Language Impairment and Nonspecific Language Impairment at Ages 4 and 6 Years Across Phenotypes of Speech, Language, and Nonverbal Cognition." Journal of Speech, Language, and Hearing Research 63, no. 3 (March 23, 2020): 793–813. http://dx.doi.org/10.1044/2019_jslhr-19-00012.
Full textAbstract:
Purpose Early language and speech acquisition can be delayed in twin children, a twinning effect that diminishes between 4 and 6 years of age in a population-based sample. The purposes of this study were to examine how twinning effects influence the identification of children with language impairments at 4 and 6 years of age, comparing children with specific language impairment (SLI) and nonspecific language impairment (NLI); the likelihood that affectedness will be shared within monozygotic versus dizygotic twin pairs; and estimated levels of heritability for SLI and NLI. Twinning effects are predicted to result in elevated rates of language impairments in twins. Method The population-based twin sample included 1,354 children from 677 twin pairs, 214 monozygotic and 463 dizygotic, enrolled in a longitudinal study. Nine phenotypes from the same comprehensive direct behavioral assessment protocol were investigated at 4 and 6 years of age. Twinning effects were estimated for each phenotype at each age using structural equation models estimated via diagonally weighted least squares. Heritabilities were calculated for SLI and NLI. Results As predicted, the twinning effect increased the percentage of affected children in both groups across multiple language phenotypes, an effect that diminished with age yet was still not aligned to singleton age peers. Substantial heritability estimates replicated across language phenotypes and increased with age, even with the most lenient definition of affectedness, at −1 SD . Patterns of outcomes differed between SLI and NLI groups. Conclusions Nonverbal IQ is not on the same causal pathway as language impairments. Twinning effects on language acquisition affect classification of 4- and 6-year-old children as SLI and NLI, and heritability is most consistent in the SLI group. Clinical practice requires monitoring language acquisition of twins to avoid misdiagnosis when young or a missed diagnosis of language impairments at school entry.
29
MALMASI, SHERVIN, and MARK DRAS. "Multilingual native language identification." Natural Language Engineering 23, no. 2 (December 2, 2015): 163–215. http://dx.doi.org/10.1017/s1351324915000406.
Full textAbstract:
AbstractWe present the first comprehensive study of Native Language Identification (NLI) applied to text written in languages other than English, using data from six languages. NLI is the task of predicting an author’s first language using only their writings in a second language, with applications in Second Language Acquisition and forensic linguistics. Most research to date has focused on English but there is a need to apply NLI to other languages, not only to gauge its applicability but also to aid in teaching research for other emerging languages. With this goal, we identify six typologically very different sources of non-English second language data and conduct six experiments using a set of commonly used features. Our first two experiments evaluate our features and corpora, showing that the features perform well and at similar rates across languages. The third experiment compares non-native and native control data, showing that they can be discerned with 95 per cent accuracy. Our fourth experiment provides a cross-linguistic assessment of how the degree of syntactic data encoded in part-of-speech tags affects their efficiency as classification features, finding that most differences between first language groups lie in the ordering of the most basic word categories. We also tackle two questions that have not previously been addressed for NLI. Other work in NLI has shown that ensembles of classifiers over feature types work well and in our final experiment we use such an oracle classifier to derive an upper limit for classification accuracy with our feature set. We also present an analysis examining feature diversity, aiming to estimate the degree of overlap and complementarity between our chosen features employing an association measure for binary data. Finally, we conclude with a general discussion and outline directions for future work.
30
Peters, Jurriaan M., Deborah P. Waber, Gloria B. McAnulty, and Frank H. Duffy. "Event-Related Correlations in Learning Impaired Children during a Hybrid Go/No-Go Choice Reaction Visual-Motor Task." Clinical Electroencephalography 34, no. 3 (July 2003): 99–109. http://dx.doi.org/10.1177/155005940303400304.
Full textAbstract:
One hundred sixty-nine learning impaired (LI) and 71 non-learning impaired (NLI) children underwent a hybrid go/no-go choice reaction time visual-motor task to study the behavioral and physiological fundamentals of learning disorders. A left button was pressed for Left Arrow (LA) stimuli, a right for Right Arrow (RA) stimuli, none (no-go) for a non-directional arrow. Stimulus specific visual evoked potentials were formed and, with PZ as index electrode, were lag-correlated to frontal electrodes to form Event-Related Correlations (ERC). Exploratory t-statistic significance probability maps (t-SPM) were used to define regions of interest (ROI). Behaviorally, there was a right-hand advantage over the left in the NLI group, but less in the LI group. Electrophysiologically, RA and LA conditions increased correlation between visual areas (PZ) and contralateral frontal areas (F3 and F4). A unilateral ROI, at electrode FC1, also preceded both left- and right-handed responses. Neurobehaviorally, increased visual-motor correlation was associated with better performance, especially for the left hemisphere, at F3 and FC1. Surprisingly, visual-motor correlations were not associated with performance for the NLI group in the RA and no-go condition. Our data support previously reported difficulties of learning impaired children in low-level information processing. Furthermore, we hypothesize that LI, in contrast to NLI children, demonstrate difficulty in automatizing routine tasks.
31
Williams, Alisha, Margaret Pulsifer, Kristin Tissera, and Leila A. Mankarious. "Cognitive and Behavioral Functioning in Hearing-Impaired Children with and without Language Delay." Otolaryngology–Head and Neck Surgery 163, no. 3 (April 14, 2020): 588–90. http://dx.doi.org/10.1177/0194599820915741.
Full textAbstract:
Poor language development in patients with sensorineural hearing loss (SNHL) may be related to an auditory deficit and/or other neurologic condition that influences the ability to communicate. A retrospective chart review of children (mean age = 4.0 years) with congenital, bilateral SNHL was performed to assess for linguistic and nonlinguistic neurodevelopmental differences between those who were language-impaired (LI) versus non–language-impaired (NLI). Language, neurodevelopmental functioning, and behavior were assessed. Twenty-two patients were identified: 12 were LI and 10 were NLI. Average pure-tone thresholds and nonverbal intelligence were not different between the language groups, but the LI group demonstrated significantly lower median overall adaptive skills, personal living skills, and motor skills. Behavioral dysregulation was significantly higher in the LI versus NLI group (58% vs 10%; P = .031), although the median neurodevelopmental scores did not differ significantly. These findings introduce the possibility that nonlinguistic processing deficit(s) may be confounding the ability to develop language.
32
Muszala, Stefan P., Gita Alaghband, James Hack, and Daniel Connors. "Natural Load Indices (NLI) for scientific simulation." Journal of Supercomputing 59, no. 1 (May 26, 2010): 392–413. http://dx.doi.org/10.1007/s11227-010-0442-y.
Full text
33
Liang, Ze, Feili Wei, Yueyao Wang, Jiao Huang, Hong Jiang, Fuyue Sun, and Shuangcheng Li. "The Context-Dependent Effect of Urban Form on Air Pollution: A Panel Data Analysis." Remote Sensing 12, no. 11 (June 2, 2020): 1793. http://dx.doi.org/10.3390/rs12111793.
Full textAbstract:
There have been debates and a lack of understanding about the complex effects of urban-scale urban form on air pollution. Based on the remotely sensed data of 150 cities in the Beijing-Tianjin-Hebei agglomeration in China from 2000 to 2015, we studied the effects of urban form on fine particulate matter (PM2.5) and nitrogen dioxide (NO2) concentrations from multiple perspectives. The panel models show that the elastic coefficients of aggregation index and fractal dimension are the highest among all factors for the whole region. Population density, aggregation index, and fractal dimension have stronger influences on air pollution in small cities, while area size demonstrates the opposite effect. Population density has a stronger impact on medium/high-elevation cities, while night light intensity (NLI), fractal dimension, and area size show the opposite effect. Low road network density can enlarge the influence magnitude of NLI and population density. The results of the linear regression model with multiplicative interactions provide evidence of interactions between population density and NLI or aggregation index. The slope of the line that captures the relationship between NLI on PM2.5 is positive at low levels of population density, flat at medium levels of population density, and negative at high levels of population density. The study results also show that when increasing the population density, the air pollution in a city with low economic and low morphological aggregation degrees will be impacted more greatly.
34
Boldeanu, Lucia-Camelia, Marius Boariu, Darian Rusu, Adrian Vaduva, Alexandra Roman, Petra Surlin, Ioana Martu, Razvan Dragoi, Aurel Popa-Wagner, and Stefan-Ioan Stratul. "Histomorphometrical and CBCT Evaluation of Tissue Loss Progression Induced by Consecutive, Alternate Ligatures in Experimental Peri-Implantitis in a Dog Model: A Pilot Study." Journal of Clinical Medicine 11, no. 20 (October 20, 2022): 6188. http://dx.doi.org/10.3390/jcm11206188.
Full textAbstract:
Objectives: Soft and hard tissue breakdown was histologically and radiologically assessed around implants with alternate, consecutively placed ligatures on the same edentulous dog hemimandible. The influence of ligatured implants (LI) on adjacent non-ligatured implants (NLI, as a possible naturally induced peri-implantitis) was also evaluated. Material and Methods: Three months after tooth extraction, five dental implants were placed in the dog hemimandible. Two months after abutment placement, ligatures were placed subsequently two months apart on alternate implants, while both intermediate implants were left without ligatures. Ligatures were kept in place during the entire experiment, and no plaque control measures were taken. Eleven months post-implantation, the animal was sacrificed. Undecalcified ground sections were cut, stained with Masson Goldner and MOVAT Pentachrome and evaluated by light microscopy. Soft and hard tissue loss was assessed using histomorphometric and CBCT parameters. Results: All NLI presented deep false peri-implant pockets on the oral aspect and pronounced vertical bone resorption on the buccal aspect. After 2, 4 and 6 months, during the breakdown period, more than 30% of the bone was lost in LI in all directions, while, despite immediate vicinity, NLI displayed less destruction. Intense inflammation, typical for induced peri-implantitis, was present, with similar intensity in LI as NLI, but in different parts of the lesions. Morphometry confirmed intense soft tissue inflammation, more bone resorption and higher amounts of infiltrated connective tissue in LI when compared with NLI. Conclusion: Within the limits of the present pilot study, the adequacy of the experimental dog model based on ligature-induced peri-implantitis was able to be successfully challenged by non-ligature models of spontaneously occurring peri-implant inflammation, while meeting the requirements for experimental designs with a very small numbers of animals. The influence of implants with severe peri-implantitis on adjacent implants resulted in less than expected tissue loss in the latter accession numbers.
35
Guo, Maosheng, Yu Zhang, and Ting Liu. "Gaussian Transformer: A Lightweight Approach for Natural Language Inference." Proceedings of the AAAI Conference on Artificial Intelligence 33 (July 17, 2019): 6489–96. http://dx.doi.org/10.1609/aaai.v33i01.33016489.
Full textAbstract:
Natural Language Inference (NLI) is an active research area, where numerous approaches based on recurrent neural networks (RNNs), convolutional neural networks (CNNs), and self-attention networks (SANs) has been proposed. Although obtaining impressive performance, previous recurrent approaches are hard to train in parallel; convolutional models tend to cost more parameters, while self-attention networks are not good at capturing local dependency of texts. To address this problem, we introduce a Gaussian prior to selfattention mechanism, for better modeling the local structure of sentences. Then we propose an efficient RNN/CNN-free architecture named Gaussian Transformer for NLI, which consists of encoding blocks modeling both local and global dependency, high-order interaction blocks collecting the evidence of multi-step inference, and a lightweight comparison block saving lots of parameters. Experiments show that our model achieves new state-of-the-art performance on both SNLI and MultiNLI benchmarks with significantly fewer parameters and considerably less training time. Besides, evaluation using the Hard NLI datasets demonstrates that our approach is less affected by the undesirable annotation artifacts.
36
Stacey, Joe, Yonatan Belinkov, and Marek Rei. "Supervising Model Attention with Human Explanations for Robust Natural Language Inference." Proceedings of the AAAI Conference on Artificial Intelligence 36, no. 10 (June 28, 2022): 11349–57. http://dx.doi.org/10.1609/aaai.v36i10.21386.
Full textAbstract:
Natural Language Inference (NLI) models are known to learn from biases and artefacts within their training data, impacting how well they generalise to other unseen datasets. Existing de-biasing approaches focus on preventing the models from learning these biases, which can result in restrictive models and lower performance. We instead investigate teaching the model how a human would approach the NLI task, in order to learn features that will generalise better to previously unseen examples. Using natural language explanations, we supervise the model’s attention weights to encourage more attention to be paid to the words present in the explanations, significantly improving model performance. Our experiments show that the in-distribution improvements of this method are also accompanied by out-of-distribution improvements, with the supervised models learning from features that generalise better to other NLI datasets. Analysis of the model indicates that human explanations encourage increased attention on the important words, with more attention paid to words in the premise and less attention paid to punctuation and stopwords.
37
Zhao, Zizhen, Chen Fu, Yuping Zhang, and Ailing Fu. "Dimeric Histidine as a Novel Free Radical Scavenger Alleviates Non-Alcoholic Liver Injury." Antioxidants 10, no. 10 (September 27, 2021): 1529. http://dx.doi.org/10.3390/antiox10101529.
Full textAbstract:
Non-alcoholic liver injury (NLI) is a common disease worldwide. Since free radical damage in the liver is a crucial initiator leading to diseases, scavenging excess free radicals has become an essential therapeutic strategy. To enhance the antioxidant capacity of histidine, we synthesized a protonated dimeric histidine, H-bihistidine, and investigated its anti-free radical potential in several free-radical-induced NLI. Results showed that H-bihistidine could strongly scavenge free radicals caused by H2O2, fatty acid, and CCl4, respectively, and recover cell viability in cultured hepatocytes. In the animal model of nonalcoholic fatty liver injury caused by high-fat diet, H-bihistidine reduced the contents of transaminases and lipids in serum, eliminated the liver’s fat accumulation, and decreased the oxidative damage. Moreover, H-bihistidine could rescue CCl4-induced liver injury and recover energy supply through scavenging free radicals. Moreover, liver fibrosis prepared by high-fat diet and CCl4 administration was significantly alleviated after H-bihistidine treatment. This study suggests a novel nonenzymatic free radical scavenger against NLI and, potentially, other free-radical-induced diseases.
38
Walden, Thomas P., Shajaky Parameswaran, Louise Brisbois, and B. Catharine Craven. "Post Doc Competition (Knowledge Generation) ID 1987817." Topics in Spinal Cord Injury Rehabilitation 29, suppl (September 1, 2023): 207. http://dx.doi.org/10.46292/sci23-1987817s.
Full textAbstract:
Background Individuals with a neurologic level of injury (NLI) C1-T10 have an UMN bowel and propensity for external anal sphincter spasm and need chemical/mechanical stimuli to evacuate their bowels versus those with an NLI T11-S5 and LMN bowel whom have a patulous sphincter and complete manual bowel disimpaction. We describe the associations between independence in sphincter control for 5 years following discharge among individuals with traumatic SCI (tSCI). Methods Adults with tSCI (n=113, 80 men) whom completed baseline, 1, 2, and 5-year community follow-up interviews (2014-2021) were included. Responses to the Spinal Cord Independence Measure (SCIM-III) Sphincter Control and Respiration subscores were extracted. Participants were separated into UMN and LMN groups. A linear mixed model determined longitudinal differences in SCIM subscores. NLI and follow-up time points were assigned as fixed effects, age a random effect, and sex as a covariate. Observed differences were compared to the minimal clinically importance difference (MCID) in SCIM-III subscores. Results The mean difference in SCIM-III subscores was 4.85 between the UMN and LMN bowel groups, with UMN group scoring lower at all-time points (p≤0.017). The mean group difference was equal to the MCID, constituting a substantial meaningful difference. No differences in SCIM subscores were noted across time (p≥0.9), regardless of NLI or bowel impairment. Conclusion Self-reported sphincter control remains stable for 5 years post-injury, emphasizing the need to maximizing bowel independence before discharge. Individuals with UMN bowel have lower sphincter control scores than individuals with a LMN bowel.
39
Kargon, Jeremy. "Under Construction: Alternative Spaces of Discourse at the National Library of Israel (Critical Essay)." Review of Middle East Studies 53, no. 01 (June 2019): 59–82. http://dx.doi.org/10.1017/rms.2019.4.
Full textAbstract:
AbstractA new building for the National Library of Israel (NLI), scheduled for completion in 2021, is the culmination of a two-decade process of institutional transformation. Formerly known as the Jewish National and University Library, the NLI has historically served simultaneously as Israel's official state repository, as the Hebrew University's central library, and as a “library of the Jewish people.” Like other national libraries around the world, including elsewhere in the Middle East, the National Library of Israel has had to grapple with accelerated changes in management of library collections due to the proliferation of digital media. More fundamental, however, have been changes in the cultural expectations about how libraries should function. Since 1998, the NLI has sought to expand its mission to promote not only scholarship but also cultural “discourse” among Israel's diverse constituencies. The architectural design of NLI's new edifice was intended, therefore, to do more than house the functional requirements of a modern library. It was commissioned to express through its design the significance of the transformed institution for the Israeli public. Towards that goal, a highly publicized competition for the NLI's design was held in 2012. The original two-stage competition ended in controversy after the architect endorsed by the jurors was dismissed. Yet a review of designs submitted by four Israeli architects in that first competition shows how public spaces, affiliated with public institutions, are expected to foster public discourse in Israel. Whether that discourse is cultural or political, contentious or contradictory, these alternative designs for the NLI illustrate common themes based upon specific environmental tropes, familiar across a broad spectrum of Israeli society.
40
Galashev, Alexander, and Alexey Vorob'ev. "An Ab Initio Study of Lithization of Two-Dimensional Silicon–Carbon Anode Material for Lithium-Ion Batteries." Materials 14, no. 21 (November 4, 2021): 6649. http://dx.doi.org/10.3390/ma14216649.
Full textAbstract:
This work is devoted to a first-principles study of changes in the structural, energetic, and electronic properties of silicene anodes during their lithium filling. Anodes were presented by silicene on carbon substrate and free-standing silicene. The ratio of the amount of lithium to silicon varied in the range from 0.06 to 1.125 for silicene on bilayer graphene and from 0.06 to 2.375 for free-standing silicene. It is shown that the carbon substrate reduces the stability of the silicene sheet. Silicene begins to degrade when the ratio of lithium to silicon (NLi/NSi) exceeds ~0.87, and at NLi/NSi = 0.938, lithium penetrates into the space between the silicene sheet and the carbon substrate. At certain values of the Li/Si ratio in the silicene sheet, five- and seven-membered rings of Si atoms can be formed on the carbon substrate. The presence of two-layer graphene imparts conductive properties to the anode. These properties can periodically disappear during the adsorption of lithium in the absence of a carbon substrate. Free-standing silicene adsorbed by lithium loses its stability at NLi/NSi = 1.375.
41
Jiang, Nan-Jiang, and Marie-Catherine de Marneffe. "Investigating Reasons for Disagreement in Natural Language Inference." Transactions of the Association for Computational Linguistics 10 (2022): 1357–74. http://dx.doi.org/10.1162/tacl_a_00523.
Full textAbstract:
Abstract We investigate how disagreement in natural language inference (NLI) annotation arises. We developed a taxonomy of disagreement sources with 10 categories spanning 3 high- level classes. We found that some disagreements are due to uncertainty in the sentence meaning, others to annotator biases and task artifacts, leading to different interpretations of the label distribution. We explore two modeling approaches for detecting items with potential disagreement: a 4-way classification with a “Complicated” label in addition to the three standard NLI labels, and a multilabel classification approach. We found that the multilabel classification is more expressive and gives better recall of the possible interpretations in the data.
42
Han, Ji-yoon. "A Basic Study of NLI Dataset for Korean." HAN-GEUL 81, no. 4 (December 31, 2020): 949–75. http://dx.doi.org/10.22557/hg.2020.12.81.4.949.
Full text
43
Talman, Aarne, Anssi Yli-Jyrä, and Jörg Tiedemann. "Sentence embeddings in NLI with iterative refinement encoders." Natural Language Engineering 25, no. 4 (July 2019): 467–82. http://dx.doi.org/10.1017/s1351324919000202.
Full textAbstract:
AbstractSentence-level representations are necessary for various natural language processing tasks. Recurrent neural networks have proven to be very effective in learning distributed representations and can be trained efficiently on natural language inference tasks. We build on top of one such model and propose a hierarchy of bidirectional LSTM and max pooling layers that implements an iterative refinement strategy and yields state of the art results on the SciTail dataset as well as strong results for Stanford Natural Language Inference and Multi-Genre Natural Language Inference. We can show that the sentence embeddings learned in this way can be utilized in a wide variety of transfer learning tasks, outperforming InferSent on 7 out of 10 and SkipThought on 8 out of 9 SentEval sentence embedding evaluation tasks. Furthermore, our model beats the InferSent model in 8 out of 10 recently published SentEval probing tasks designed to evaluate sentence embeddings’ ability to capture some of the important linguistic properties of sentences.
44
Rothbart, Zack. "COVID-19 crisis response at the National Library of Israel: Confronting challenges and maximising opportunities." Alexandria: The Journal of National and International Library and Information Issues 30, no. 2-3 (August 2020): 126–34. http://dx.doi.org/10.1177/0955749020980140.
Full textAbstract:
Founded in 1892, the National Library of Israel (NLI) serves as the vibrant institution of national memory for the Jewish people worldwide and Israelis of all backgrounds and faiths. Its four core collections – Israel, Judaica, Islam and Middle East, and the Humanities – tell the historical, cultural and intellectual story of the Jewish people, the State of Israel and the Land of Israel and its region throughout the ages. The NLI’s current transformative renewal aims to encourage diverse audiences in Israel and across the globe to engage with these treasures in meaningful ways through a range of innovative educational, cultural and digital initiatives. The most tangible manifestation of this transformation is the new NLI campus, now under construction adjacent to the Knesset (Israeli Parliament) in Jerusalem, and on schedule to open its doors in 2022. NLI’s renewal and dual mandate requiring it to engage diverse domestic and international audiences, as well as the massive construction project underway, have in many ways magnified the challenges posed by this difficult period, as well as – and perhaps even more so – the opportunities it presents. While the response to these unprecedented and unforeseen circumstances has largely been ad hoc, the NLI approach has been guided by the goal of protecting the health and welfare of its staff and users, and identifying strategic opportunities to not only make the most of the difficulties presented by this complex new reality but also build programs and initiatives to help achieve strategic goals. Following a brief summary of the crisis in Israel, this article presents a number of examples of the physical, logistical and programmatic adaptations NLI has implemented in attempting to maximise potential opportunities in best fulfilling its mission during this time.
45
Liga, Gabriele, Astrid Barreiro, Hami Rabbani, and Alex Alvarado. "Extending Fibre Nonlinear Interference Power Modelling to Account for General Dual-Polarisation 4D Modulation Formats." Entropy 22, no. 11 (November 20, 2020): 1324. http://dx.doi.org/10.3390/e22111324.
Full textAbstract:
In optical communications, four-dimensional (4D) modulation formats encode information onto the quadrature components of two arbitrary orthogonal states of polarisation of the optical field. Many analytical models available in the optical communication literature allow, within a first-order perturbation framework, the computation of the average power of the nonlinear interference (NLI) accumulated in coherent fibre-optic transmission systems. However, all such models only operate under the assumption of transmitted polarisation-multiplexed two-dimensional (PM-2D) modulation formats, which only represent a limited subset of the possible dual-polarisation 4D (DP-4D) formats. Namely, only those where data transmitted on each polarisation channel are mutually independent and identically distributed. This paper presents a step-by-step mathematical derivation of the extension of existing NLI models to the class of arbitrary DP-4D modulation formats. In particular, the methodology adopted follows the one of the popular enhanced Gaussian noise model, albeit dropping most assumptions on the geometry and statistic of the transmitted 4D modulation format. The resulting expressions show that, whilst in the PM-2D case the NLI power depends only on different statistical high-order moments of each polarisation component, for a general DP-4D constellation, several other cross-polarisation correlations also need to be taken into account.
46
Khan, Asmat Ullah, Rasheed Ahmad Khera, Naveed Anjum, Rao Aqil Shehzad, Saleem Iqbal, Khurshid Ayub, and Javed Iqbal. "DFT study of superhalogen and superalkali doped graphitic carbon nitride and its non-linear optical properties." RSC Advances 11, no. 14 (2021): 7779–89. http://dx.doi.org/10.1039/d0ra08608h.
Full text
47
Braschler, Lorin, Ulrich Kraus, Thomas Braschler, and Beat Knechtle. "Skisturz mit temporärer Tetraparese." Praxis 111, no. 13 (October 2022): 760–65. http://dx.doi.org/10.1024/1661-8157/a003937.
Full textAbstract:
Zusammenfassung. Wir präsentieren einen 74-jährigen Patienten nach Skisturz und danach aufgetretener inkompletter, armbetonter sensibler und motorisch spastischer Tetraparese. Bei Vorstellung in der Hausarztpraxis konnten sowohl Pyramidenbahnzeichen sowie Störungen der Feinmotorik in beiden Händen beobachtet werden. Zudem zeigte sich ein sensibles und ataktisches Gangbild. Klinisch zeigte sich zusammenfassend eine traumatische inkomplette sensible und motorische Querschnittslähmung NLI C5 und AIS D. Die MRT-Untersuchung der Halswirbelsäule zeigte passend hierzu eine hochgradige Spinalkanalstenose auf Höhe C5 mit Myelonkompression. Es erfolgte eine operative Dekompression des Rückenmarks mit anschliessender Fusion und Stabilisierung der entsprechenden Halswirbelkörper. Postoperativ zeigte sich klinisch weiterhin eine Querschnittslähmung NLI C5 und AIS D. Nach dreiwöchiger neurologischer Rehabilitation hat sich der Patient bis auf eine leichte Paraspastik der Beine und noch leicht bestehenden Hypästhesien der Fingerbeeren erholt.
48
Lahey, Margaret, and Jan Edwards. "Why Do Children With Specific Language Impairment Name Pictures More Slowly Than Their Peers?" Journal of Speech, Language, and Hearing Research 39, no. 5 (October 1996): 1081–98. http://dx.doi.org/10.1044/jshr.3905.1081.
Full textAbstract:
To examine the role of different cognitive processes in accounting for the slower naming times of children with specific language impairment (SLI) relative to peers with no language impairment (NLI), three tasks designed to stress different types of processing were administered: naming pictures with the signal to respond presented at various delay intervals, naming following different durations of exposure to identical and unrelated primes, and vocally responding to nonlinguistic stimuli. Children with SLI, aged 4 to 9.5 years, were significantly slower than their NLI age peers on naming and on responding to nonlinguistic stimuli, but the effect of delay interval before naming and of duration of prime exposure before naming was similar for both groups. Results suggested that speed of naming is related to the slower nonlinguistic response processing of children with SLI and not to speed of their linguistic or perceptual processing. To examine differences in processing that might relate to pattern of language performance we examined responses of two subgroups of SLI. The subgroup of children whose language problems involved expressive but not receptive skills was not significantly slower than their NLI peers. The children whose problems involved both expressive and receptive language were significantly slower, but this was influenced by age. Findings are discussed in terms of language performance, age, task variables, and a generalized rate-limiting factor.
49
Ghosh, Srutarshi, Anil Kumar Gupta, Dileep Kumar, Sudhir Mishra, Ganesh Yadav, and Avinash Agarwal. "Pulmonary Function Improves in Persons with Paraplegia after Partial Body Weight Supported Treadmill Training: a Prospective Randomized Study." Central European Journal of Sport Sciences and Medicine 41 (2023): 49–60. http://dx.doi.org/10.18276/cej.2023.1-05.
Full textAbstract:
Objectives: To evaluate changes in Pulmonary Function Test (PFT) parameters in individuals with paraplegia following Partial Body Weight Supported Treadmill Training (PBWSTT).
Design: Randomized controlled trial
Setting: Inpatient rehabilitation facility
Participants: Adults with chronic SCI (n = 42).
Intervention: Patients were randomly allocated in CR group (N= 20) receiving Conventional Rehabilitation or in PBWSTT group (N=22) receiving both Conventional Rehabilitation and PBWSTT for 4 weeks.
Main outcome measure(s): Changes in % predicted PFT parameter for the subject’s age, sex and BMI.
Results: With PBWSTT, significant PFT changes were VC (P =.009), PEF (p = .001) and ERV (p = .032). In complete SCI, PEF (p = .026) improved, while in incomplete SCI VC (p = .005), ERV (p = .029), PEF( p = .001) improved with PBWSTT. In upper neurological level of injury (NLI) (T6-T11), PBWSTT improved PEF (p = .004) alone while in lower NLI (T12-L2), with PBWSTT both ERV (p = .016) and PEF (p = .035) improved.
Conclusions: With added PBWSTT most parameters including Vital Capacity, the global measure of PFT, improved significantly, especially in Lower NLI and incomplete SCI. The positive role of this noninvasive exercise based intervention in improving lung functions comes as an added benefit to the usual benefit of locomotion. This may encourage researchers to design future larger studies to validate it aiming the inclusion of PBWSTT in routine SCI rehabilitation protocols.
50
Shang, Mingyue, Zhenxin Fu, Hongzhi Yin, Bo Tang, Dongyan Zhao, and Rui Yan. "Find a Reasonable Ending for Stories: Does Logic Relation Help the Story Cloze Test?" Proceedings of the AAAI Conference on Artificial Intelligence 33 (July 17, 2019): 10031–32. http://dx.doi.org/10.1609/aaai.v33i01.330110031.
Full textAbstract:
Natural language understanding is a challenging problem that covers a wide range of tasks. While previous methods generally train each task separately, we consider combining the cross-task features to enhance the task performance. In this paper, we incorporate the logic information with the help of the Natural Language Inference (NLI) task to the Story Cloze Test (SCT). Previous work on SCT considered various semantic information, such as sentiment and topic, but lack the logic information between sentences which is an essential element of stories. Thus we propose to extract the logic information during the course of the story to improve the understanding of the whole story. The logic information is modeled with the help of the NLI task. Experimental results prove the strength of the logic information.