Hasil untuk "Chinese language and literature"

Tran Sy Bao

Sign language translation systems typically require English as an intermediary language, creating barriers for non-English speakers in the global deaf community. We present Canonical Semantic Form (CSF), a language-agnostic semantic representation framework that enables direct translation from any source language to sign language without English mediation. CSF decomposes utterances into nine universal semantic slots: event, intent, time, condition, agent, object, location, purpose, and modifier. A key contribution is our comprehensive condition taxonomy comprising 35 condition types across eight semantic categories, enabling nuanced representation of conditional expressions common in everyday communication. We train a lightweight transformer-based extractor (0.74 MB) that achieves 99.03% average slot extraction accuracy across four typologically diverse languages: English, Vietnamese, Japanese, and French. The model demonstrates particularly strong performance on condition classification (99.4% accuracy) despite the 35-class complexity. With inference latency of 3.02ms on CPU, our approach enables real-time sign language generation in browser-based applications. We release our code, trained models, and multilingual dataset to support further research in accessible sign language technology.

Daniel Gallagher, Gerhard Heyer

This paper evaluates the performance of transformer-based language models on split-ergative case alignment in Georgian, a particularly rare system for assigning grammatical cases to mark argument roles. We focus on subject and object marking determined through various permutations of nominative, ergative, and dative noun forms. A treebank-based approach for the generation of minimal pairs using the Grew query language is implemented. We create a dataset of 370 syntactic tests made up of seven tasks containing 50-70 samples each, where three noun forms are tested in any given sample. Five encoder- and two decoder-only models are evaluated with word- and/or sentence-level accuracy metrics. Regardless of the specific syntactic makeup, models performed worst in assigning the ergative case correctly and strongest in assigning the nominative case correctly. Performance correlated with the overall frequency distribution of the three forms (NOM > DAT > ERG). Though data scarcity is a known issue for low-resource languages, we show that the highly specific role of the ergative along with a lack of available training data likely contributes to poor performance on this case. The dataset is made publicly available and the methodology provides an interesting avenue for future syntactic evaluations of languages where benchmarks are limited.

Krasimir Kabakciev

Compositional aspect is an extremely important cross-language and universal language phenomenon discovered by Henk Verkuyl, a Dutch linguist and logician, in 1971, published a year later. Compositional aspect appears to exist in all natural languages, and proof is being accumulated lately in favor of this thesis. In contrast to verbal aspect as found in the Slavic languages, Greek, Georgian, Chinese, etc., where it resides in verbs as lexical entries but is grammaticalized, compositional aspect exists at the level of the sentence. The paper discusses some major notions that, on the one hand, underlie the correct understanding of compositional aspect and, on the other, constitute at least some of the reasons for the absence of the article-aspect interplay, an important phenomenon, in English grammars and in the literature on English language teaching. Publications in these two fields are practically silent about a phenomenon that has otherwise been discussed in theoretical linguistics for decades. Theoretical linguistics itself, although cognizant of the article-aspect interplay, also abounds in misconceptualizations about aspect in general, reigning in it from as early as when compositional aspect was discovered. English grammars and the overall literature on English language teaching continue to remain silent on the article-aspect interplay even lately, despite the appearance in recent years of a couple of publications – successfully addressing the issue but isolated, including two revolutionary papers dealing with articles, determiners, quantifiers and similar elements taking part in the compositional effectuation of aspect. Disclosure Statement The author reported no potential conflicts of interest.

Gábor Péter Boros

The practice of meditation has become popular in Western countries during the last decades as an Eastern, and predominantly Buddhist, exercise. However, when properly compared, Christian contemplation and Buddhist meditation show numerous similarities. This paper sets out to determine the extent of comparability between these practices of the two traditions and how they are embedded in their respective theoretical frameworks. Additionally, as a further step, it aims to demonstrate how contemplative practice can act as a basis for Buddhist–Christian dialogue. The research centres around the works of two eminent monks of the 20thcentury, the American Trappist Thomas Merton and the Vietnamese Thiền master Thích Nhất Hạnh. Their work provides an ideal target of analysis, since both were not only widely appreciated members of their respective traditions but also deeply engaged in interreligious dialogue, and thus they had a proper understanding of each other’s tradition. In this paper, the author will first analyse and compare the works of Nhất Hạnh and Merton written on contemplation during approximately the same time to determine what exactly they understand by the terms ‘meditation’ and ‘contemplative prayer’, respectively. Then, their works regarding the other monk’s respective religious tradition will be discussed and compared to see what the two authors select as main avenues of comparison. The main conclusions of the article are that the concepts of meditation and contemplation in Buddhism and Christianity denote strikingly similar exercises, which aim to achieve similar goals despite differences in theoretical formulation. For Merton, Christian contemplation aims to reconnect the believer to God with whom the initial unity was lost through the Fall of Man and the establishment of the ego-self as an entity separate from God. Such contemplation needs to happen in interior solitude and involve a complete self-emptying of the believer to become one with God. Since separation from God through the ego is the fundamental problem of humans, reconnecting to God through contemplation is seen as the highest form of prayer and, indeed, life. In a similar vein, for Nhất Hạnh, the fundamental problem of humanity is the misconception of a permanent self, which results in dividing the world into different, separate entities. Thus, for him, similarly to Merton, the goal is to reach a supreme unity beyond distinctions and duality. Moreover, the way to such unity is precisely meditation, which can best unify body and mind and thus bring forth Buddhahood. Then, as the second step, the article presents how Merton and Nhất Hạnh draw parallels between Christianity and Buddhism on the basis of similar contemplation. Merton points to, among others, the connection between God and the absolute Void of Zen, kenosis and śūnyatā, whereas Nhất Hạnh points to similarities between nirvāṇa and the Kingdom of God as well as mindfulness and the Holy Spirit. Thus, the paper provides a thorough analysis of the similarities of meditation practice in Christianity and Buddhism, as practiced by Thomas Merton and Thích Nhất Hạnh, and how these similarities can provide a basis for dialogue between the two religions in modern times.

Balázs Szabó

Morito Tatsuo, a Tokió Császári Egyetem adjunktusa 1920-ban az egyetem közgazdászklubjának folyóiratában cikket publikált Kropotkin társadalomfilozófiájáról címmel, amelyben ismertette a neves anarchista gondolkodó elképzeléseit az igazságos társadalomról, és ezeket megjegyzéseivel látta el. A cikk nagy port vert fel, amelyet az Igazságügyminisztérium a „társadalmi békéről szóló” törvény, illetve a sajtótörvény egyes cikkelyei alapján sértőnek talált, és végül büntetőfeljelentést tett. Morito-t nemcsak az állásából bocsájtották el, de háromhavi börtönbüntetésre is ítélték. Jelen tanulmány ennek az ügynek az előzményeit és következményeit vizsgálja. A háttér fontos részét jelenti az egyetemi közeg változása az 1910-es években: részben a háborús konjunktúrának, részben pedig a Meiji-kori hagyományok elhalványulásának köszönhetően a japán lakosság szélesebb rétegei számára vált lehetővé az egyetemi oktatás, a hallgatók felduzzadó létszáma az oktatók számának növekedését vonta maga után, mindez pedig óhatatlanul új eszmék, az alacsonyabb társadalmi sorból érkező szereplők miatt pedig a tömegek sorsa iránti nagyobb érzékenységet hozta be az akadémiai világba. Ugyanebben az időben a közgazdaság-tudománynak mint önálló diszciplínának a megjelenése és a jogi tudományoktól valófüggetlenedése zajlott le, ami ezt a tudományágat immár az államtudományoktól függetlenül rajzolta újra, és ez új kérdésfeltevésekhez és néhány kutató, mint ahogy Morito esetében, a társadalmi jólét és igazságosság problémájának a diskurzusba való bevonásához vezetett el. Ezen túl kapcsolat mutatható ki a Taishō-kor sajátos japán politikai viszonyaihoz, a megerősödő szélsőjobboldalhoz is: az ügy kirobbanásakor kormányon lévő Hara Takashi jobbközép kabinetje természetesen fordult szövetségesként a tőlük jobbra állókhoz az egyre erősödő baloldali hangulattal szemben. A Morito-ügy érdekessége, szemben az 1905-ös Tomizu-esettel, amikor még a Kiotói Egyetem is a Tōdai mellé állt, hogy nem vezetett el az akadémiai világ összefogásához, de köze volt a Tōdai közgazdasági tanszékén kialakuló frakcióharcokhoz, amelyek majd az 1930-as években botrányok sorát okozták. A tanulmány megpróbálja bemutatni a Taishō-időszak egyetemi világát és annak társadalmi, politikai kapcsolatait is, egy olyan akadémiai közeget, amely már elmozdult a Meiji-kor hagyományaitól, kereste a helyét a japán társadalomban, és ahol a kutatói szabadság, egyetemi autonómia kérdései újra és újra az általában vett szabadság és véleményszabadság problémáival találkoztak.

Meihua Liu, Tianhao Li

Background: Second language (L2) learning is complex, multifaceted, and greatly influenced by various factors, of which individual factors like anxiety and motivation are important ones. Though anxiety and motivation have been shown to be strongly correlated with each other and interact with other variables to collaboratively affect L2 learning, mixed findings have been revealed, demonstrating the complexity of the interrelationship between L2 anxiety and motivation and their interactive effects on L2 learning. Purpose: Guided by the self-determination theory, this study aimed to explore the levels of and the relationship between English classroom anxiety and motivation as well as their predictive effects on Chinese undergraduate EFL (English as a foreign language) learners’ English achievement. Method: The participants were 571 Chinese university students who answered an 8-item English Classroom Anxiety Scale, a 35-item English Learning Motivation Scale, and a 5-item Demographic Information Questionnaire. They also reported their scores in tests that they had recently taken and self-rated their overall English proficiency as indicators of their English achievement. Results: The study revealed the following major findings: (a) the participants had a small to moderate level of English classroom anxiety, and a medium level of intrinsic motivation, extrinsic motivation, personal goals and expectancy/control in English learning, (b) English classroom anxiety was significantly negatively correlated with all motivation scales, (c) English classroom anxiety was not only significantly negatively related to but negatively predicted the students’ English achievement, and (d) significantly positive correlations existed between English learning motivation and English achievement. Of different motivation dimensions, intrinsic and extrinsic motivation positively predicted the latter. Conclusion: The findings of this study further demonstrate the importance of anxiety and motivation in L2 learning and the need to explore anxiety-reduction strategies, increase students’ intrinsic and extrinsic motivation, and strengthen their expectancy in L2 teaching and learning, thus contributing to the understanding of foreign language anxiety and L2 motivation and enriching the current literature on the two issues.

Xinyu Zhang, Xinyu Zhang, Lingyong Xiao et al.

BackgroundDiabetic peripheral neuropathy (DPN) is one of the most common complications of diabetes mellitus, often causing pain or numbness in the patient’s limbs and even leading to amputation and death. Elderly patients with DPN usually have higher morbidity and more severe results. Acupuncture has been widely used as an effective treatment for DPN in China. However, the efficacy of acupuncture in the treatment of DPN remains unclear. In this review, we aimed to explore the impact of acupuncture in alleviating symptoms of DPN.Method and analysisSix databases were searched from inception to October 2023. We searched Medline, EMBASE, Cochrane Central Register of Controlled Trials (CENTRAL), and three Chinese databases, namely China National Knowledge Infrastructure (CNKI), SinoMed, and Wanfang. All randomized controlled trials related to the effect of acupuncture on DPN will be included. There was no restriction in language or publication year. The primary outcome is the response rate. The secondary outcomes are the Toronto clinical scoring system (TCSS), nerve conduction velocities (NCVs), and blood glucose before and after the treatment. Two researchers will be responsible for the selection of study, data extraction, and assessment of study quality independently. RevMan V5.1.0 software will be used to assess the risk of bias and generate data.ResultsWe searched 4518 studies, among which 9 RCTs were considered eligible. Overall, acupuncture treatment had a higher response rate than controls (relative risk (RR), −2.87 [95% confidence interval (CI), −5.27 to −0.48], p = 0.02) and significantly alleviated the symptoms of DPN patients, reduced their blood glucose levels, and improved their NCVs compared to the control group. This study will provide a high-quality synthesis of current available evidence for the clinical treatment of DPN with this therapy.ConclusionThe results suggested that acupuncture might be effective in improving symptoms of DPN in elderly patients. Owing to the overall low quality of the literature included, we need more large-sample, high-quality, and low-bias studies to prove it.

Jayanta Sadhu, Maneesha Rani Saha, Rifat Shahriyar

The rapid growth of Large Language Models (LLMs) has put forward the study of biases as a crucial field. It is important to assess the influence of different types of biases embedded in LLMs to ensure fair use in sensitive fields. Although there have been extensive works on bias assessment in English, such efforts are rare and scarce for a major language like Bangla. In this work, we examine two types of social biases in LLM generated outputs for Bangla language. Our main contributions in this work are: (1) bias studies on two different social biases for Bangla, (2) a curated dataset for bias measurement benchmarking and (3) testing two different probing techniques for bias detection in the context of Bangla. This is the first work of such kind involving bias assessment of LLMs for Bangla to the best of our knowledge. All our code and resources are publicly available for the progress of bias related research in Bangla NLP.

Sourojit Ghosh, Srishti Chatterjee

This chapter focuses on gender-related errors in machine translation (MT) in the context of low-resource languages. We begin by explaining what low-resource languages are, examining the inseparable social and computational factors that create such linguistic hierarchies. We demonstrate through a case study of our mother tongue Bengali, a global language spoken by almost 300 million people but still classified as low-resource, how gender is assumed and inferred in translations to and from the high(est)-resource English when no such information is provided in source texts. We discuss the postcolonial and societal impacts of such errors leading to linguistic erasure and representational harms, and conclude by discussing potential solutions towards uplifting languages by providing them more agency in MT conversations.

Sebastian Pagel, Michael Jirasek, Leroy Cronin

Chemputation is the process of programming chemical robots to do experiments using a universal symbolic language, but the literature can be error prone and hard to read due to ambiguities. Large Language Models (LLMs) have demonstrated remarkable capabilities in various domains, including natural language processing, robotic control, and more recently, chemistry. Despite significant advancements in standardizing the reporting and collection of synthetic chemistry data, the automatic reproduction of reported syntheses remains a labour-intensive task. In this work, we introduce an LLM-based chemical research agent workflow designed for the automatic validation of synthetic literature procedures. Our workflow can autonomously extract synthetic procedures and analytical data from extensive documents, translate these procedures into universal XDL code, simulate the execution of the procedure in a hardware-specific setup, and ultimately execute the procedure on an XDL-controlled robotic system for synthetic chemistry. This demonstrates the potential of LLM-based workflows for autonomous chemical synthesis with Chemputers. Due to the abstraction of XDL this approach is safe, secure, and scalable since hallucinations will not be chemputable and the XDL can be both verified and encrypted. Unlike previous efforts, which either addressed only a limited portion of the workflow, relied on inflexible hard-coded rules, or lacked validation in physical systems, our approach provides four realistic examples of syntheses directly executed from synthetic literature. We anticipate that our workflow will significantly enhance automation in robotically driven synthetic chemistry research, streamline data extraction, improve the reproducibility, scalability, and safety of synthetic and experimental chemistry.

Shujuan Zhao, Lingfeng Qiao, Kangyang Luo et al.

Large language models (LLMs) have become powerful tools for advancing natural language processing applications in the financial industry. However, existing financial LLMs often face challenges such as hallucinations or superficial parameter training, resulting in suboptimal performance, particularly in financial computing and machine reading comprehension (MRC). To address these issues, we propose a novel large language model specifically designed for the Chinese financial domain, named SNFinLLM. SNFinLLM excels in domain-specific tasks such as answering questions, summarizing financial research reports, analyzing sentiment, and executing financial calculations. We then perform the supervised fine-tuning (SFT) to enhance the model's proficiency across various financial domains. Specifically, we gather extensive financial data and create a high-quality instruction dataset composed of news articles, professional papers, and research reports of finance domain. Utilizing both domain-specific and general datasets, we proceed with continuous pre-training on an established open-source base model, resulting in SNFinLLM-base. Following this, we engage in supervised fine-tuning (SFT) to bolster the model's capability across multiple financial tasks. Crucially, we employ a straightforward Direct Preference Optimization (DPO) method to better align the model with human preferences. Extensive experiments conducted on finance benchmarks and our evaluation dataset demonstrate that SNFinLLM markedly outperforms other state-of-the-art financial language models. For more details, check out our demo video here: https://www.youtube.com/watch?v=GYT-65HZwus.

Tianmai M. Zhang, Neil F. Abernethy

Reference errors, such as citation and quotation errors, are common in scientific papers. Such errors can result in the propagation of inaccurate information, but are difficult and time-consuming to detect, posing a significant challenge to scientific publishing. To support automatic detection of reference errors, this work evaluated the ability of large language models in OpenAI's GPT family to detect quotation errors. Specifically, we prepared an expert-annotated, general-domain dataset of statement-reference pairs from journal articles. Large language models were evaluated in different settings with varying amounts of reference information provided by retrieval augmentation. Our results showed that large language models are able to detect erroneous citations with limited context and without fine-tuning. This study contributes to the growing literature that seeks to utilize artificial intelligence to assist in the writing, reviewing, and publishing of scientific papers. Potential avenues for further improvements in this task are also discussed.

Liu Dingfei

The Critical Period Hypothesis is still controversial recently, so it cannot be said to be absolutely confirmed to be false. It is more important to realize the significance of application. Although the exact properties of this phenomenon remain unclear and the literature is divided, more and more Chinese mothers choose to teach English or other languages as a second language to their children from a very early age (0-3 years old). This paper discusses whether these attempts can help children achieve the future goals set by their parents through literature review. The results attempt to support the existence of the Critical Period Hypothesis, because from the end of adolescence, the ability to learn grammar declines sharply, and the ability to reach native-like proficiency depends largely on the age at which learning begins.

Jiahao Liang, Yuyu Zhang, Penghao Chi et al.

Background: Essential oils (EOs) are a plant-derived volatile mixture. Due to their extensive biological activity, EOs have been utilized as ancient remedies to treat a variety of illnesses. Objectives: Our review aims to expand our understanding of EOs’ biological activity on the central nervous system (CNS) and to highlight the importance of utilizing EOs in CNS disorders. Method: A systematic literature search was conducted using PubMed, ScienceDirect, China National Knowledge Infrastructure (CNKI), Springer Link, Wan-fang database, and Chinese Biomedicine Database (CBM). The search was conducted to collect relevant journals and contents using the following terms: ''essential oils'', ''aromatherapy'', ''essential oils'' and ''CNS''. Language of publications was unlimited. Results: EOs are concentrated volatile aromatic liquids extracted from natural plants with different chemical compositions, obtained by physical and chemical methods such as distillation and pressing. This paper explores the effects of EOs products on CNS including promoting intelligence, improving sleep, promoting cognition and memory, anti-anxiety and depression, sedation and anti-epilepsy. EOs exert their neuropharmacological effects through blood circulation or the olfactory system. This paper summarizes clinical studies showing that aromatherapy could improve sleep, relieve preoperative anxiety and postoperative PONV, relieve gynecological disease pain, and play a role in hospice care. Conclusion: The present findings suggest that EOs have neuropharmacological effects such as nootropic, sleep improvement, anti-dementia, anti-anxiety and depression, analgesic effect, and antiepileptic, emphasizing the importance of EOs in CNS disorders and indicating the potential clinical application.

Violita Shafira, Chen Rui

This research is entitled Representation of Chinese Culture in Wish Dragon Movie. Using semiotic analysis according to Roland Barthes. Semiotic analysis is used to analyze the signs that represent culture in the film Wish Dragon. This study aims to find the existence of a sign and then find out how Chinese culture is represented through the film Wish Dragon Movie by examining the meaning of denotation, connotation and myth according to Roland Barthes' semiotic model. In addition to finding out how the foreign audience acceptance to the Wish Dragon film and how the foreign audience understands the Chinese culture in this film, the researcher using audience comments in the most popular rated movies website to analyze.

Tim Schopf, Karim Arabi, Florian Matthes

As an efficient approach to understand, generate, and process natural language texts, research in natural language processing (NLP) has exhibited a rapid spread and wide adoption in recent years. Given the increasing research work in this area, several NLP-related approaches have been surveyed in the research community. However, a comprehensive study that categorizes established topics, identifies trends, and outlines areas for future research remains absent. Contributing to closing this gap, we have systematically classified and analyzed research papers in the ACL Anthology. As a result, we present a structured overview of the research landscape, provide a taxonomy of fields of study in NLP, analyze recent developments in NLP, summarize our findings, and highlight directions for future work.

Guoxing Yang, Jianyu Shi, Zan Wang et al.

Pre-training and fine-tuning have emerged as a promising paradigm across various natural language processing (NLP) tasks. The effectiveness of pretrained large language models (LLM) has witnessed further enhancement, holding potential for applications in the field of medicine, particularly in the context of Traditional Chinese Medicine (TCM). However, the application of these general models to specific domains often yields suboptimal results, primarily due to challenges like lack of domain knowledge, unique objectives, and computational efficiency. Furthermore, their effectiveness in specialized domains, such as Traditional Chinese Medicine, requires comprehensive evaluation. To address the above issues, we propose a novel domain specific TCMDA (TCM Domain Adaptation) approach, efficient pre-training with domain-specific corpus. Specifically, we first construct a large TCM-specific corpus, TCM-Corpus-1B, by identifying domain keywords and retreving from general corpus. Then, our TCMDA leverages the LoRA which freezes the pretrained model's weights and uses rank decomposition matrices to efficiently train specific dense layers for pre-training and fine-tuning, efficiently aligning the model with TCM-related tasks, namely TCM-GPT-7B. We further conducted extensive experiments on two TCM tasks, including TCM examination and TCM diagnosis. TCM-GPT-7B archived the best performance across both datasets, outperforming other models by relative increments of 17% and 12% in accuracy, respectively. To the best of our knowledge, our study represents the pioneering validation of domain adaptation of a large language model with 7 billion parameters in TCM domain. We will release both TCMCorpus-1B and TCM-GPT-7B model once accepted to facilitate interdisciplinary development in TCM and NLP, serving as the foundation for further study.

Ainayya Almira, Anisah Rachmawati, Insi Norma Jelita et al.

The aim of this research is to provide insight to chemistry education teachers and researchers regarding the effectiveness of the guided inquiry learning model and provide direction for further research in this field. The research method used in this article is Systematic Literature Review (SLR), to help compile and evaluate various research related to the guided inquiry learning model. The instrument used in this research is to present the results of a literature review of various articles discussing the application of this model in chemistry learning by exploring the definition, application, strengths, weaknesses and effectiveness of the guided inquiry learning model in chemistry learning. The research results show that the application of this model can be carried out both in the theoretical and practical aspects of chemistry learning. The advantages of the guided inquiry model involve students actively, increase learning independence, and provide students with the opportunity to discuss and find their own answers. Students who study with this model tend to have higher learning achievements. However, there are also disadvantages, such as the time required to implement this model and obstacles in dealing with students who are not yet familiar with this approach.

Xianghui Sun, Yunjie Ji, Baochang Ma et al.

Recently, the instruction-tuning of large language models is a crucial area of research in the field of natural language processing. Due to resource and cost limitations, several researchers have employed parameter-efficient tuning techniques, such as LoRA, for instruction tuning, and have obtained encouraging results In comparison to full-parameter fine-tuning, LoRA-based tuning demonstrates salient benefits in terms of training costs. In this study, we undertook experimental comparisons between full-parameter fine-tuning and LoRA-based tuning methods, utilizing LLaMA as the base model. The experimental results show that the selection of the foundational model, training dataset scale, learnable parameter quantity, and model training cost are all important factors. We hope that the experimental conclusions of this paper can provide inspiration for training large language models, especially in the field of Chinese, and help researchers find a better trade-off strategy between training cost and model performance. To facilitate the reproduction of the paper's results, the dataset, model and code will be released.

Jia-bin Wang, Yan-xia Wang, Fen Li et al.

Objective: To assess the efficacy of Da Chaihu decoction combined with metformin tablets on patients with type 2 diabetes compared with metformin alone. Methods: This systematic review and meta‐analysis is written based on 2020 PRISMA Extension for Chinese Herbal Medicines 2020 (PRISMA-CHM 2020) reporting guidelines. We reviewed all the relevant studies from a search of the following databases from inception to February 2022 without any language restriction: Excerpta Medica Database (EMBASE), Google Scholar, PubMed, Cochrane Library, China National Knowledge Infrastructure (CNKI), VIP Information, Wanfang Data, and the Chinese Biomedical Literature Database(CBM). Data were extracted and the quality was independently evaluated by two reviewers, based on the inclusion and exclusion criteria. Data were analyzed using the Cochrane software RevMan 5.3. Results: Six randomized controlled trials comprising 516 participants were included.The meta‐analysis revealed the Da Chaihu decoction combined with metformin tablets group was significantly superior to the metformin tablets group in terms of fasting blood glucose(FPG) (−0.66 mmol/L; 95 % CI (confidence intervals) [− 1.28, − 0.04]), plasma glucose 2 h after meal (2-h PG) (−1.18 mmol/L; 95 % CI [−1.94, −0.42]) in six RCTs, body mass index (BMI) (−3.07 mmol/L; 95 % CI [−6.89, 0.75]) in three RCTs, glycosylated hemoglobin (HbAlc) (−0.36 mmol/L; 95 % CI [−1.04, 0.31]) in three RCTs, and triglycerides (TG) (−0.76 mmol/L; 95 % CI [−1.37, −0.15]) in two RCTs. In two RCTs, there were significant differences in terms of total cholesterol (TC) (−0.97 mmol/L; 95 % CI [−1.18, −0.76]). Conclusions: Very low-quality research shows that Da Chaihu decoction combined with metformin tablets exert a certain level of efficacy on patients with type 2 diabetes compared with metformin alone. However, random sequence generation methodology was reported in five studies leading to the low quality of the included studies. None of the six studies depicted the blinding method, allocation concealment, selective reporting, and assessed the purity and potency of the product. This observation requires verification through high-quality, multi-center, double-blinded randomized controlled trials, and assesses the purity and potency of the product.