$f \propto r^{-α} \cdot (r+γ)^{-β}$ has been empirically shown more precise than a naïve power law $f\propto r^{-α}$ to model the rank-frequency ($r$-$f$) relation of words in natural languages. This work shows that the only crucial parameter in the formulation is $γ$, which depicts the resistance to vocabulary growth on a corpus. A method of parameter estimation by searching an optimal $γ$ is proposed, where a ``zeroth word'' is introduced technically for the calculation. The formulation and parameters are further discussed with several case studies.
This paper presents an in-depth analysis of the latency characteristics observed in simultaneous speech-to-speech model's performance, particularly focusing on hallucination-induced latency spikes. By systematically experimenting with various input parameters and conditions, we propose methods to minimize latency spikes and improve overall performance. The findings suggest that a combination of careful input management and strategic parameter adjustments can significantly enhance speech-to-speech model's latency behavior.
SplaXBERT, built on ALBERT-xlarge with context-splitting and mixed precision training, achieves high efficiency in question-answering tasks on lengthy texts. Tested on SQuAD v1.1, it attains an Exact Match of 85.95% and an F1 Score of 92.97%, outperforming traditional BERT-based models in both accuracy and resource efficiency.
AbstractThis study reports on the intrinsic and fast self‐healing ability of all inorganic, low‐toxic Cs2ZnX4 (X = Cl, Br) metal halide nanoparticles (NPs) when subjected to local heating by electron beam irradiation in high‐resolution transmission electron microscopy (HR‐TEM). The local heating induces the creation of nanoshells (NSs) following the template of the corresponding NPs, which are subsequently healed back to their original state within several minutes. Energy dispersive spectroscopy (EDS) and fast Fourier transform (FFT) analysis reveal that the composition, phase, and crystallographic structure of the original NPs are restored during the self‐healing process, with a thin crystalline layer observed at the bottom of the NSs acting as the healing template. The inelastic scattering of the electron beam energy generates local heat that causes rapid atomic displacement, resulting in atomic mobility that lowers the density of the material and leads to NS formation. A unique insitu TEM heating stage measurement demonstrates the appearance of identical damage and self‐healing to those induced by the electron beam. The NPs exhibit excellent stability under ambient conditions for up to a month, making them suitable for self‐healing scintillators and other optoelectronic applications that require atomic‐scale stability and healing.
This paper describes an approach to detect idiomaticity only from the contextualized representation of a MWE over multilingual pretrained language models. Our experiments find that larger models are usually more effective in idiomaticity detection. However, using a higher layer of the model may not guarantee a better performance. In multilingual scenarios, the convergence of different languages are not consistent and rich-resource languages have big advantages over other languages.
Anastassia Kornilova, Daniel Argyle, Vladimir Eidelman
In this paper, we apply Item Response Theory, popular in education and political science research, to the analysis of argument persuasiveness in language. We empirically evaluate the model's performance on three datasets, including a novel dataset in the area of political advocacy. We show the advantages of separating these components under several style and content representations, including evaluating the ability of the speaker embeddings generated by the model to parallel real-world observations about persuadability.
Automatic summarization is the process of shortening a set of textual data computationally, to create a subset (a summary) that represents the most important pieces of information in the original text. Existing summarization methods can be roughly divided into two types: extractive and abstractive. An extractive summarizer explicitly selects text snippets (words, phrases, sentences, etc.) from the source document, while an abstractive summarizer generates novel text snippets to convey the most salient concepts prevalent in the source.
Recent techniques for the task of short text clustering often rely on word embeddings as a transfer learning component. This paper shows that sentence vector representations from Transformers in conjunction with different clustering methods can be successfully applied to address the task. Furthermore, we demonstrate that the algorithm of enhancement of clustering via iterative classification can further improve initial clustering performance with different classifiers, including those based on pre-trained Transformer language models.
This study aims at solving the Machine Reading Comprehension problem where questions have to be answered given a context passage. The challenge is to develop a computationally faster model which will have improved inference time. State of the art in many natural language understanding tasks, BERT model, has been used and knowledge distillation method has been applied to train two smaller models. The developed models are compared with other models which have been developed with the same intention.
This paper proposes a method for classifying movie genres by only looking at text reviews. The data used are from Large Movie Review Dataset v1.0 and IMDb. This paper compared a K-nearest neighbors (KNN) model and a multilayer perceptron (MLP) that uses tf-idf as input features. The paper also discusses different evaluation metrics used when doing multi-label classification. For the data used in this research, the KNN model performed the best with an accuracy of 55.4\% and a Hamming loss of 0.047.
This paper describes a builder entry, named "strawman", to the sentence-level sentiment analysis task of the "Build It, Break It" shared task of the First Workshop on Building Linguistically Generalizable NLP Systems. The goal of a builder is to provide an automated sentiment analyzer that would serve as a target for breakers whose goal is to find pairs of minimally-differing sentences that break the analyzer.
In this paper, we proposed two strategies which can be applied to a multilingual neural machine translation system in order to better tackle zero-shot scenarios despite not having any parallel corpus. The experiments show that they are effective in terms of both performance and computing resources, especially in multilingual translation of unbalanced data in real zero-resourced condition when they alleviate the language bias problem.
This article provides an interesting exploration of character-level convolutional neural network solving Chinese corpus text classification problem. We constructed a large-scale Chinese language dataset, and the result shows that character-level convolutional neural network works better on Chinese corpus than its corresponding pinyin format dataset. This is the first time that character-level convolutional neural network applied to text classification problem.
Representation learning is the dominant technique for unsupervised domain adaptation, but existing approaches often require the specification of "pivot features" that generalize across domains, which are selected by task-specific heuristics. We show that a novel but simple feature embedding approach provides better performance, by exploiting the feature template structure common in NLP problems.
ARKref is a tool for noun phrase coreference. It is a deterministic, rule-based system that uses syntactic information from a constituent parser, and semantic information from an entity recognition component. Its architecture is based on the work of Haghighi and Klein (2009). ARKref was originally written in 2009. At the time of writing, the last released version was in March 2011. This document describes that version, which is open-source and publicly available at: http://www.ark.cs.cmu.edu/ARKref
In the geolocation field where high-level programs and low-level devices coexist, it is often difficult to find a friendly user inter- face to configure all the parameters. The challenge addressed in this paper is to propose intuitive and simple, thus natural lan- guage interfaces to interact with low-level devices. Such inter- faces contain natural language processing and fuzzy represen- tations of words that facilitate the elicitation of business-level objectives in our context.