TasTe: Teaching Large Language Models to Translate through Self-Reflection

Machine translation aims to minimize the language barrier between people of different linguistic backgrounds. Machine translation is an automatic translation technique between pairs of different languages. Machine translation using neural network or neural machine translation came into the picture due to several limitation associated with its predecessors which are rule based and statistical based models. For large data sets with a rich range of vocabulary, the neural network machine translation system provides fair translation accuracy. We have observed that there remain very few machine translators dedicated to Indian languages, especially those spoken in the North-East area. So our paper mainly focuses on implementing a neural machine translation system for the English-Assamese language. In this paper we used five different neural machine translation models for English-Assamese language pair using LSTMs and GRUs, along with an attention layer. We have made a comparison analysis based on the performance results of these models. We used BLEU Score to calculate the accuracy of these five models, thus achieving a BLEU score of 34.168%, which was the highest among the five models.

The question of context in neural machine translation often focuses on topics related to document-level translation or intersentential context. However, there is a wide range of other aspects that can be considered under the umbrella of context. In this work, we survey ways that researchers have incorporated context into neural machine translation systems and the evaluation thereof. This includes building translation systems that operate at the paragraph level or the document level or ones that translate at the sentence level but incorporate information from other sentences. We also consider how issues like terminology consistency, anaphora, and world knowledge or external information can be considered as types of context relevant to the task of machine translation and its evaluation. Closely tied to these topics is the question of how to best evaluate machine translation output in a way that is sensitive to the contexts in which it appears. To this end, we discuss work on incorporating context into both human and automatic evaluations of machine translation quality. Furthermore, we also discuss recent experiments in the field as they relate to the use of large language models in translation and evaluation. We conclude with a view of the future of machine translation, where we expect to see issues of context continue to come to the forefront.

Machine Translation (MT) systems are now being improved with the use of an ongoing methodology known as Neural Machine Translation (NMT). Natural language processing (NLP) researchers have shown that NMT systems are unable to deal with out-of-vocabulary (OOV) words and multi-word expressions (MWEs) in the text. OOV terms are those that are not currently included in the vocabulary that is used by the NMT system. MWEs are phrases that consist of a minimum of two terms but are treated as a single unit. MWEs have great importance in NLP, linguistic theory, and MT systems. In this article, OOV words and MWEs are handled for the Punjabi to English NMT system. A parallel corpus for Punjabi to English containing MWEs was developed and used to train the different models of NMT. Punjabi is a low-resource language as it lacks the availability of a large parallel corpus for building various NLP tools, and this is an attempt to improve the accuracy of Punjabi in the English NMT system by using named entities and MWEs in the corpus. The developed NMT models were assessed using human evaluation through adequacy, fluency and overall rating as well as automated assessment tools such as the bilingual evaluation study (BLEU) and translation error rate (TER) score. Results show that using word embedding (WE) and MWEs corpus increased the accuracy of translation for the Punjabi to English language pair. The best BLEU score obtained was 15.45 for the small test set, 43.32 for the medium test set, and 34.5 for the large test set, respectively. The best TER rate score obtained was 57.34% for the small test set, 37.29% for the medium test set, and 53.79% for the large test set, repectively.

With the active participation of users in product review platforms, online consumer-generated content, and, more specifically, user-generated reviews, have become a clear reference in purchasing decision-making processes, which sometimes exceed the impact of advertising campaigns. A common feature of most tourism review platforms is the use of machine translation (MT) systems to immediately make reviews available to users in various languages. However, the quality of the MT output of these reviews varies greatly, primarily due to the subjective and unstructured nature of this digital genre. Different studies confirm that there are no universal quality rating scales. The assessment of MT output quality usually depends on factors such as the purpose of the text or the value given to the immediacy of the translation. New neural MT systems have been a revolution in the quality increase of the translated output; however, new lines of research are opening up to verify whether the quality of this new paradigm of MT can be assessed with the existing scales, mainly from previous rule-based systems and statistical translation, or whether it is necessary to develop new quality metrics specifically for these new intelligent systems. On the other hand, one of the questions that remain to be resolved in this new context of neural MT is whether the use of large amounts of textual data in the training of these systems is as effective as the use of less data but of higher quality and better-adjusted to the specialty and type of text for which it is used. Based on the hypothesis that each genre requires specific quality rating scales, this work identifies the error patterns and textual characteristics of online user reviews from a corpus-based approach analysis that will contribute to adapting quality rating scales to this specific digital genre.

Translating culture-related content is vital for effective cross-cultural communication.However, many culture-specific items (CSIs) often lack viable translations across languages, making it challenging to collect high-quality, diverse parallel corpora with CSI annotations.This difficulty hinders the analysis of cultural awareness of machine translation (MT) systems, including traditional neural MT and the emerging MT paradigm using large language models (LLM).To address this gap, we introduce a novel parallel corpus, enriched with CSI annotations in 6 language pairs for investigating Culturally-Aware Machine Translation-CAMT. 1 Furthermore, we design two evaluation metrics to assess CSI translations, focusing on their pragmatic translation quality.Our findings show the superior ability of LLMs over neural MTs in leveraging external cultural knowledge for translating CSIs, especially those lacking translations in the target culture.

State-of-the-art (SOTA) neural machine translation (NMT) systems translate texts at sentence level, ignoring context: intra-textual information, like the previous sentence, and extra-textual information, like the gender of the speaker. As a result, some sentences are translated incorrectly. Personalised NMT (PersNMT) and document-level NMT (DocNMT) incorporate this information into the translation process. Both fields are relatively new and previous work within them is limited. Moreover, there are no readily available robust evaluation metrics for them, which makes it difficult to develop better systems, as well as track global progress and compare different methods. This thesis proposal focuses on PersNMT and DocNMT for the domain of dialogue extracted from TV subtitles in five languages: English, Brazilian Portuguese, German, French and Polish. Three main challenges are addressed: (1) incorporating extra-textual information directly into NMT systems; (2) improving the machine translation of cohesion devices; (3) reliable evaluation for PersNMT and DocNMT.

The preservation of domain knowledge from source to the target is crucial in any translation workflows. Hence, translation service providers that use machine translation (MT) in production could reasonably expect that the translation process should transfer both the underlying pragmatics and the semantics of the source-side sentences into the target language. However, recent studies suggest that the MT systems often fail to preserve such crucial information (e.g., sentiment, emotion, gender traits) embedded in the source text in the target. In this context, the raw automatic translations are often directly fed to other natural language processing (NLP) applications (e.g., sentiment classifier) in a cross-lingual platform. Hence, the loss of such crucial information during the translation could negatively affect the performance of such downstream NLP tasks that heavily rely on the output of the MT systems. In our current research, we carefully balance both the sides (i.e., sentiment and semantics) during translation, by controlling a global-attention-based neural MT (NMT), to generate translations that encode the underlying sentiment of a source sentence while preserving its non-opinionated semantic content. Toward this, we use a state-of-the-art reinforcement learning method, namely, actor-critic , that includes a novel reward combination module, to fine-tune the NMT system so that it learns to generate translations that are best suited for a downstream task, viz. sentiment classification while ensuring the source-side semantics is intact in the process. Experimental results for Hindi–English language pair show that our proposed method significantly improves the performance of the sentiment classifier and alongside results in an improved NMT system.

Introduction to the special issue on deep learning approaches for machine translation

Extracting information from unstructured and semi-structured geoscience literature is a crucial step in conducting geological research. The traditional machine learning extraction paradigm requires a substantial amount of high-quality manually annotated data for model training, which is time-consuming, labor-intensive, and not easily transferable to new fields. Recently, large language models (LLMs) (e.g., ChatGPT, GPT-4, and LLaMA), have shown great performance in various natural language processing (NLP) tasks, such as question answering, machine translation, and text generation. A substantial body of work has demonstrated that LLMs possess strong in-context learning (ICL) and even zero-shot learning capabilities to solve downstream tasks without specifically designed supervised fine-tuning. In this paper, we propose utilizing LLMs for geoscience literature information extraction. Specifically, we design a hierarchical PDF parsing pipeline and an automated knowledge extraction process, which can significantly reduce the need for manual data annotation, assisting geoscientists in literature data mining. For the hierarchical PDF parsing pipeline, firstly, a document layout detection model fine-tuned on geoscience literature is employed for layout detection, obtaining layout detection information for the document. Secondly, based on the document layout information, an optical character content parsing model is used for content parsing, obtaining the text structure and plain text corresponding to the content. Finally, the text structure and plain text are combined and reconstructed to ultimately obtain the parsed structured data. For the automated knowledge extraction process, firstly, the parsed long text is segmented into paragraphs to adapt to the input length limit of LLMs. Subsequently, a few-shot prompting method is employed for structured knowledge extraction, encompassing two tasks: attribute value extraction and triplet extraction. In attribute value extraction, prompts are generated automatically by the LLMs based on the subdomain and attribute names, facilitating the location and extraction of values related to subdomain attribute names in the text. For triplet extraction, the LLMs employ a procedural approach to entity extraction, entity type extraction, and relation extraction, following the knowledge graph structure pattern. Finally, the extracted structured knowledge is stored in the form of knowledge graphs, facilitating further analysis and integration of various types of knowledge from the literature. Our proposed approach turns out to be simple, flexible, and highly effective in geoscience literature information extraction. Demonstrations of information extraction in subdomains such as radiolarian fossils and fluvial facies have yielded satisfactory results. The extraction efficiency has significantly improved, and feedback from domain experts indicates a relatively high level of accuracy in the extraction process. The extracted results can be used to construct a foundational knowledge graph for geoscience literature, supporting the comprehensive construction and efficient application of a geoscience knowledge graph.

A novel prompting method for few-shot NER via LLMs

Corpus-based machine translation (MT) has been the main approach to developing and implementing MT systems in both academia and the industry over the last three decades. In this field, the type and size of the corpus used for training MT engines have presented problems for both statistical MT (SMT) systems as well as neural MT (NMT) systems, being the two dominant corpusbased approaches. Moreover, language pairs such as Turkish-English have been understudied within this framework. This article aims to evaluate the translation quality in Turkish-to-English custom MT systems that have been trained on different corpus sizes and types. Two NMT engines and two SMT engines were trained on the KantanMT platform using two different training corpus types with either only domain-specific cardiology corpus or this corpus plus a mixed-domain corpus. The study conducted both automatic evaluations with metrics including BLEU, F-Measure and TER, as well as a comprehensive human evaluation with metrics including fluency, A/B test, and adequacy. Lastly, the study realized a separate, subjective terminology evaluation in order to investigate how differently MT systems handle terminology, as this is a crucial aspect for specific-domain text types such as cardiology. While the automatic evaluation results suggest the SMT engines to perform better than NMT engines, all human evaluators rated the mixed-domain NMT engine as the highest performing one. However, the terminology evaluation task demonstrated SMT to still be able to perform better and to commit less terminology errors, despite the industry and academia shifting toward NMT engines.

Recent advancements in Neural MachineTranslation (NMT) systems have significantly improved model performance on various translation benchmarks.However, these systems still face numerous challenges when translating low-resource languages such as Urdu.In this work, we highlight the specific issues faced by machine translation systems when translating Urdu language.We first conduct a comprehensive evaluation of English to Urdu Machine Translation with four diverse models: GPT-3.5 (a large language model), opus-mt-en-ur (a bilingual translation model), NLLB (a model trained for translating 200 languages) and IndicTrans2 (a specialized model for translating lowresource Indic languages).The results demonstrate that IndicTrans2 significantly outperforms other models in Urdu Machine Translation.To understand the differences in the performance of these models, we analyze the Urdu word distribution in different training datasets and compare the training methodologies.Finally, we uncover the specific translation issues and provide suggestions for improvements in Urdu machine translation systems.

This work aims to present an overview of using some artificial intelligence (AI) models in machine translation (MT). This work aims to integrate machine learning (ML), deep learning (DL), large language models (LLMs) to enhance machine translation between natural languages. The focus is directed to present the neural-based machine translation (NMT), and some DL models are presented. The bidirectional-encoder-representation from transformer (BERT) and LLMs are presented to utilize the big amount of textual data to learn translation patterns. The main measurable criteria that are used to evaluate the performance of MT and Arabic machine translation (AMT) are also presented. Some linguistic and technical challenges of MT and AMT are discussed. Some key points of future works in NMT are mentioned. A comparative study among some recent published related works is presented. A critical survey is presented to show the important role of DL and LLMs in MT. Some open-source toolkits, datasets and some commercial MT systems are collected and briefly presented. This work is expected to be useful for those people interested to know the up-to-date knowledge of MT and the potential role of DL and LLM in automatic translation.

Generative large language models (LLMs) are a subset of transformers-based neural network architecture models. LLMs have successfully leveraged a combination of an increased number of parameters, improvements in computational efficiency, and large pre-training datasets to perform a wide spectrum of natural language processing (NLP) tasks. Using a few examples (few-shot) or no examples (zero-shot) for prompt-tuning has enabled LLMs to achieve state-of-the-art performance in a broad range of NLP applications. This article by the American Medical Informatics Association (AMIA) NLP Working Group characterizes the opportunities, challenges, and best practices for our community to leverage and advance the integration of LLMs in downstream NLP applications effectively. This can be accomplished through a variety of approaches, including augmented prompting, instruction prompt tuning, and reinforcement learning from human feedback (RLHF). Our focus is on making LLMs accessible to the broader biomedical informatics community, including clinicians and researchers who may be unfamiliar with NLP. Additionally, NLP practitioners may gain insight from the described best practices. We focus on 3 broad categories of NLP tasks, namely natural language understanding, natural language inferencing, and natural language generation. We review the emerging trends in prompt tuning, instruction fine-tuning, and evaluation metrics used for LLMs while drawing attention to several issues that impact biomedical NLP applications, including falsehoods in generated text (confabulation/hallucinations), toxicity, and dataset contamination leading to overfitting. We also review potential approaches to address some of these current challenges in LLMs, such as chain of thought prompting, and the phenomena of emergent capabilities observed in LLMs that can be leveraged to address complex NLP challenge in biomedical applications.

BackgroundA large language model (LLM) is a machine learning model inferred from text data that captures subtle patterns of language use in context. Modern LLMs are based on neural network architectures that incorporate transformer methods. They allow the model to relate words together through attention to multiple words in a text sequence. LLMs have been shown to be highly effective for a range of tasks in natural language processing (NLP), including classification and information extraction tasks and generative applications.ObjectiveThe aim of this adapted Delphi study was to collect researchers’ opinions on how LLMs might influence health care and on the strengths, weaknesses, opportunities, and threats of LLM use in health care.MethodsWe invited researchers in the fields of health informatics, nursing informatics, and medical NLP to share their opinions on LLM use in health care. We started the first round with open questions based on our strengths, weaknesses, opportunities, and threats framework. In the second and third round, the participants scored these items.ResultsThe first, second, and third rounds had 28, 23, and 21 participants, respectively. Almost all participants (26/28, 93% in round 1 and 20/21, 95% in round 3) were affiliated with academic institutions. Agreement was reached on 103 items related to use cases, benefits, risks, reliability, adoption aspects, and the future of LLMs in health care. Participants offered several use cases, including supporting clinical tasks, documentation tasks, and medical research and education, and agreed that LLM-based systems will act as health assistants for patient education. The agreed-upon benefits included increased efficiency in data handling and extraction, improved automation of processes, improved quality of health care services and overall health outcomes, provision of personalized care, accelerated diagnosis and treatment processes, and improved interaction between patients and health care professionals. In total, 5 risks to health care in general were identified: cybersecurity breaches, the potential for patient misinformation, ethical concerns, the likelihood of biased decision-making, and the risk associated with inaccurate communication. Overconfidence in LLM-based systems was recognized as a risk to the medical profession. The 6 agreed-upon privacy risks included the use of unregulated cloud services that compromise data security, exposure of sensitive patient data, breaches of confidentiality, fraudulent use of information, vulnerabilities in data storage and communication, and inappropriate access or use of patient data.ConclusionsFuture research related to LLMs should not only focus on testing their possibilities for NLP-related tasks but also consider the workflows the models could contribute to and the requirements regarding quality, integration, and regulations needed for successful implementation in practice.