Causal reasoning aims to predict the future scenarios that may be caused by the observed actions. However, existing causal reasoning methods deal with causalities on the word level. In this paper, we propose a novel event-level causal reasoning metho
d and demonstrate its use in the task of effect generation. In particular, we structuralize the observed cause-effect event pairs into an event causality network, which describes causality dependencies. Given an input cause sentence, a causal subgraph is retrieved from the event causality network and is encoded with the graph attention mechanism, in order to support better reasoning of the potential effects. The most probable effect event is then selected from the causal subgraph and is used as guidance to generate an effect sentence. Experiments show that our method generates more reasonable effect sentences than various well-designed competitors.
Pre-trained neural language models give high performance on natural language inference (NLI) tasks. But whether they actually understand the meaning of the processed sequences is still unclear. We propose a new diagnostics test suite which allows to
assess whether a dataset constitutes a good testbed for evaluating the models' meaning understanding capabilities. We specifically apply controlled corruption transformations to widely used benchmarks (MNLI and ANLI), which involve removing entire word classes and often lead to non-sensical sentence pairs. If model accuracy on the corrupted data remains high, then the dataset is likely to contain statistical biases and artefacts that guide prediction. Inversely, a large decrease in model accuracy indicates that the original dataset provides a proper challenge to the models' reasoning capabilities. Hence, our proposed controls can serve as a crash test for developing high quality data for NLI tasks.
This paper describes models developed for the Social Media Mining for Health (SMM4H) 2021 shared tasks. Our team participated in the first subtask that classifies tweets with Adverse Drug Effect (ADE) mentions. Our best performing model utilizes BERT
weet followed by a single layer of BiLSTM. The system achieves an F-score of 0.45 on the test set without the use of any auxiliary resources such as Part-of-Speech tags, dependency tags, or knowledge from medical dictionaries.
This paper presents our contribution to the Social Media Mining for Health Applications Shared Task 2021. We addressed all the three subtasks of Task 1: Subtask A (classification of tweets containing adverse effects), Subtask B (extraction of text sp
ans containing adverse effects) and Subtask C (adverse effects resolution). We explored various pre-trained transformer-based language models and we focused on a multi-task training architecture. For the first subtask, we also applied adversarial augmentation techniques and we formed model ensembles in order to improve the robustness of the prediction. Our system ranked first at Subtask B with 0.51 F1 score, 0.514 precision and 0.514 recall. For Subtask A we obtained 0.44 F1 score, 0.49 precision and 0.39 recall and for Subtask C we obtained 0.16 F1 score with 0.16 precision and 0.17 recall.
The paper researches the problem of drug adverse effect detection in texts of social media. We describe the development of such classification system for Russian tweets. To increase the train dataset we apply a couple of augmentation techniques and analyze their effect in comparison with similar systems presented at 2021 years' SMM4H Workshop.
This paper provides a quick overview of possible methods how to detect that reference translations were actually created by post-editing an MT system. Two methods based on automatic metrics are presented: BLEU difference between the suspected MT and
some other good MT and BLEU difference using additional references. These two methods revealed a suspicion that the WMT 2020 Czech reference is based on MT. The suspicion was confirmed in a manual analysis by finding concrete proofs of the post-editing procedure in particular sentences. Finally, a typology of post-editing changes is presented where typical errors or changes made by the post-editor or errors adopted from the MT are classified.
Abstract Recent improvements in the predictive quality of natural language processing systems are often dependent on a substantial increase in the number of model parameters. This has led to various attempts of compressing such models, but existing m
ethods have not considered the differences in the predictive power of various model components or in the generalizability of the compressed models. To understand the connection between model compression and out-of-distribution generalization, we define the task of compressing language representation models such that they perform best in a domain adaptation setting. We choose to address this problem from a causal perspective, attempting to estimate the average treatment effect (ATE) of a model component, such as a single layer, on the model's predictions. Our proposed ATE-guided Model Compression scheme (AMoC), generates many model candidates, differing by the model components that were removed. Then, we select the best candidate through a stepwise regression model that utilizes the ATE to predict the expected performance on the target domain. AMoC outperforms strong baselines on dozens of domain pairs across three text classification and sequence tagging tasks.1
This study was designed to estimate the allelophatic effect of cogongrass (Imperata
cylindrica L.), in the growth of one year old of olive seedlings, in a nursery (greenhouse)
conditions. Two experiments were conducted in this study. In the first e
xperiment, the
effect of the aqueous extracts of cognograss leaves, rhizomes, and roots, at concentrations
of (2%, 4% and 8%) on the growth of olive seedlings (total length and diameter) were
evaluated. The results of this experiment indicated an inhibitory effect of the 4% and 8%
aqueous extracts on growth parameters. It was found that treatments with 4% and 8%
aqueous extracts caused significant reduction (60.2% and 83%) respectively in the total
length, in compare to the control. As for the 2% aqueous extract, it showed a stimulation
effect in the growth, an increase about 31.5% was recorded for the total length in compare
to the control. A similar result was observed in regard the seedlings diameter growth. The
2% aqueous extract showed an 56.8% increase in the rate of diameter growth, while the
4% and 8% extracts showed a reduction effects 80% and 91.5% respectively. The second
experiment evaluated the effect of dried powder of cogongrass parts at concentrations of
(2%, 4% and 8%) on the growth of olive seedlings.
This paper deals with the link between Semitic and Arabic languages,that it showed
the ancientness of the Arabic , and explained the concept of Semitic languages, and
mentioned the most important views in their home country and their common
charac
teristics. The research then presented the historical understanding of the Arab-
Semitic relationship, confirming the interaction that emerged in the pre-Islamic era through
interaction with the Bible, which was translated into Arabic
Whether the Arabs are translators or were the people of these heavenly books; the
result proves the principle of interaction between languages that these books were written
with and the Arabic language, with the continuation of this interaction clearly in the era of
the Prophet and his companions through various forms and ways .
The research then discusses the phenomenon of expression, then the ancientness of
the phenomenon of expression in Semitic languages in general, and Arabic in particular,
through the presentation of the most important proof of authenticity in these languages,
Although the Arabic is the most conservative.
The research ended with conclusions and recommendations.
The aim of this study was to study the plant species diversity at Al-Kahaf protected
area in A-Sheikh Badr region, Tartous governorate, in terms of functional
characteristics of the recorded plant species, its uses. The results can contribute
in u
nderstanding, and determining the role of those plant species for ecosystem
function and local community and help in the management of this protected area.
Plant species diversity was studied on three versants; using the intercepted line
method; and in the watercourses surrounding Al Kahaf castle. Additional surveys
have also been conducted over the whole site. Life forms, dispersal types, and the
uses of these species were recorded. The recorded species belong to 53 plant family.
Fabaceae was the most represented family (17 species), followed by Asteraceae
(12 species), and Lamiaceae (11 species).