Hasil untuk "Japanese language and literature"

Mengjie Zhao, Lianbo Liu, Yusuke Fujita et al.

SpeechLLMs typically combine ASR-trained encoders with text-based LLM backbones, leading them to inherit written-style output patterns unsuitable for text-to-speech synthesis. This mismatch is particularly pronounced in Japanese, where spoken and written registers differ substantially in politeness markers, sentence-final particles, and syntactic complexity. We propose a preference-based alignment approach to adapt Japanese SpeechLLMs for speech-worthy outputs: text that is concise, conversational, and readily synthesized as natural speech. To rigorously evaluate this task, we introduce SpokenElyza, a benchmark for Japanese speech-worthiness derived from ELYZA-tasks-100 with auditory verification by native experts. Experiments show that our approach achieves substantial improvement on SpokenElyza while largely preserving performance on the original written-style evaluation. We will release SpokenElyza to support future research on Japanese spoken dialog systems.

Yasmin Moslem, Aman Kassahun Wassie, Amanuel Gizachew Abebe

In this work, we present AfriNLLB, a series of lightweight models for efficient translation from and into African languages. AfriNLLB supports 15 language pairs (30 translation directions), including Swahili, Hausa, Yoruba, Amharic, Somali, Zulu, Lingala, Afrikaans, Wolof, and Egyptian Arabic, as well as other African Union official languages such as Arabic (MSA), French, Portuguese, and Spanish. Our training data covers bidirectional translation between English and 13 languages, and between French and two languages (Lingala and Wolof). AfriNLLB models are based on NLLB-200 600M, which we compress using iterative layer pruning and quantization. We fine-tune the pruned models on parallel corpora we curated for African languages, employing knowledge distillation from a larger teacher model. Our work aims at enabling efficient deployment of translation models for African languages in resource-constrained settings. Our evaluation results demonstrate that AfriNLLB models achieve performance comparable to the baseline while being significantly faster. We release two versions of the AfriNLLB models, a Transformers version that allows further fine-tuning and a CTranslate2 version for efficient inference. Moreover, we release all the training data that we used for fine-tuning the baseline and pruned models to facilitate further research.

Nina Cornyetz

            As translated Japanese literature crosses over to become world literature, there still are Orientalist assumptions about modern Japanese literature across the global academy at large, as of limited scope, rejecting fictionalization, and exploring states of mind. The objective of this essay is to counter essentialist and dated assumptions about modern Japanese literature by highlighting the actual breadth and diversity of English translations that negate these stereotypes. I even question postulating any homogeneous genre under the rubric “Japanese literature.” I begin with a broad survey of English-language reviews by non-area-specialists of three contemporary texts of Japanese literature in translation: Mizumura Minae’s  (b. 1951) A True Novel (Honkaku shōsetsu, 2002), Kirino Natsuo’s (b. 1951) OUT (AUTO, 1997), and Kaneshiro Kazuki’s (b. 1968) Go (Gō, 2000).  I follow that survey by revisiting and putting into question the dominant literary discourse by Japan specialists regarding just what constitutes “modernity” in Japanese literary studies. I then return my focus to the three contemporary Japanese novels already introduced. In the following order I will take up the complexities of each of the three novels’ plots, narration strategies, focalization, issues of ethnicity and race, relation of the individual to social conflicts and issues, and degrees of fictionalization versus realism. Finally, I will show how these various aspects of each of the three can moreover be seen as complementary to some of the most highly regarded fictions of the belle lettres traditions of Japanese modern literature, and hence belong in an alternative genealogy of modern Japanese literature.

Yui Fukuda, Atsuko Taguchi, Akane Futami et al.

Abstract Background Foreign workers are at risk for depression, and Vietnamese people tend to be reluctant to seek professional mental health care. Although Vietnamese people are the largest population among foreign workers in Japan, evidence concerning their help-seeking experiences and strategies to promote help-seeking in this population is lacking. This study aimed to identify the percentage of Vietnamese migrant workers in Japan who have sought help from healthcare professionals for depressive symptoms and to explore the factors related to their intentions to seek help from a psychiatrist. Methods An online questionnaire was administered to Vietnamese migrants working in Japan from October 5, 2021, to November 1, 2021. Patient Health Questionnaire-9 (PHQ-9) scores were calculated to measure the severity of the respondents’ depression. Help-seeking experiences related to depressive symptoms were also investigated. The General Help-Seeking Questionnaire Vignette Version (GHSQ-V) was modified and adapted to measure the respondents’ intention to seek help from a psychiatrist for depressive symptoms. To investigate the factors related to help-seeking intention, potential factors were selected from literature reviews and discussions with professionals in the field. Descriptive statistics were calculated, and multiple logistic regression analysis was conducted. Results A total of 803 eligible data points were collected. Among the 53.5% of participants who scored 10 or more on the PHQ-9, 4.4% had sought help from a healthcare professional in Japan. A lower preference for coping on one’s own, greater recognition of the effectiveness of help, greater ability to make work adjustments, and a higher level of Japanese language proficiency were related to greater intentions to seek help from a psychiatrist. Conclusions Many Vietnamese migrant workers in Japan are unable to seek help from healthcare professionals for depressive symptoms. It may be beneficial to acknowledge not only linguistic barriers but also other related factors when planning strategies to enhance Vietnamese migrant workers’ intentions to seek help from psychiatrists.

Yumou Wei, Paulo Carvalho, John Stamper

GPT has become nearly synonymous with large language models (LLMs), an increasingly popular term in AIED proceedings. A simple keyword-based search reveals that 61% of the 76 long and short papers presented at AIED 2024 describe novel solutions using LLMs to address some of the long-standing challenges in education, and 43% specifically mention GPT. Although LLMs pioneered by GPT create exciting opportunities to strengthen the impact of AI on education, we argue that the field's predominant focus on GPT and other resource-intensive LLMs (with more than 10B parameters) risks neglecting the potential impact that small language models (SLMs) can make in providing resource-constrained institutions with equitable and affordable access to high-quality AI tools. Supported by positive results on knowledge component (KC) discovery, a critical challenge in AIED, we demonstrate that SLMs such as Phi-2 can produce an effective solution without elaborate prompting strategies. Hence, we call for more attention to developing SLM-based AIED approaches.

Jose F Quesada

While language technologies have advanced significantly, current approaches fail to address the complex sociocultural dimensions of linguistic preservation. AI Thinking proposes a meaning-centered framework that would transform technological development from creating tools FOR communities to co-creating solutions WITH them. This approach recognizes that meaningful solutions emerge through the interplay of cultural understanding, community agency, and technological innovation. The proposal articulates a holistic methodology and a five-layer technological ecosystem where communities maintain control over their linguistic and cultural knowledge representation. This systematic integration of community needs, cultural preservation, and advanced capabilities could revolutionize how we approach linguistic diversity preservation in the digital age.

Giorgio Roffo

This tutorial explores the advancements and challenges in the development of Large Language Models (LLMs) such as ChatGPT and Gemini. It addresses inherent limitations like temporal knowledge cutoffs, mathematical inaccuracies, and the generation of incorrect information, proposing solutions like Retrieval Augmented Generation (RAG), Program-Aided Language Models (PAL), and frameworks such as ReAct and LangChain. The integration of these techniques enhances LLM performance and reliability, especially in multi-step reasoning and complex task execution. The paper also covers fine-tuning strategies, including instruction fine-tuning, parameter-efficient methods like LoRA, and Reinforcement Learning from Human Feedback (RLHF) as well as Reinforced Self-Training (ReST). Additionally, it provides a comprehensive survey of transformer architectures and training techniques for LLMs. The source code can be accessed by contacting the author via email for a request.

Mingjie Sun, Xinlei Chen, J. Zico Kolter et al.

We observe an empirical phenomenon in Large Language Models (LLMs) -- very few activations exhibit significantly larger values than others (e.g., 100,000 times larger). We call them massive activations. First, we demonstrate the widespread existence of massive activations across various LLMs and characterize their locations. Second, we find their values largely stay constant regardless of the input, and they function as indispensable bias terms in LLMs. Third, these massive activations lead to the concentration of attention probabilities to their corresponding tokens, and further, implicit bias terms in the self-attention output. Last, we also study massive activations in Vision Transformers. Code is available at https://github.com/locuslab/massive-activations.

Miguel Rios

Large Language Models (LLMs) have shown promising results on machine translation for high resource language pairs and domains. However, in specialised domains (e.g. medical) LLMs have shown lower performance compared to standard neural machine translation models. The consistency in the machine translation of terminology is crucial for users, researchers, and translators in specialised domains. In this study, we compare the performance between baseline LLMs and instruction-tuned LLMs in the medical domain. In addition, we introduce terminology from specialised medical dictionaries into the instruction formatted datasets for fine-tuning LLMs. The instruction-tuned LLMs significantly outperform the baseline models with automatic metrics.

Benjamin Clavié

Neural Information Retrieval has advanced rapidly in high-resource languages, but progress in lower-resource ones such as Japanese has been hindered by data scarcity, among other challenges. Consequently, multilingual models have dominated Japanese retrieval, despite their computational inefficiencies and inability to capture linguistic nuances. While recent multi-vector monolingual models like JaColBERT have narrowed this gap, they still lag behind multilingual methods in large-scale evaluations. This work addresses the suboptimal training methods of multi-vector retrievers in lower-resource settings, focusing on Japanese. We systematically evaluate and improve key aspects of the inference and training settings of JaColBERT, and more broadly, multi-vector models. We further enhance performance through a novel checkpoint merging step, showcasing it to be an effective way of combining the benefits of fine-tuning with the generalization capabilities of the original checkpoint. Building on our analysis, we introduce a novel training recipe, resulting in the JaColBERTv2.5 model. JaColBERTv2.5, with only 110 million parameters and trained in under 15 hours on 4 A100 GPUs, significantly outperforms all existing methods across all common benchmarks, reaching an average score of 0.754, significantly above the previous best of 0.720. To support future research, we make our final models, intermediate checkpoints and all data used publicly available.

Leonardo Ranaldi, Andrè Freitas

The alignments of reasoning abilities between smaller and larger Language Models are largely conducted via Supervised Fine-Tuning (SFT) using demonstrations generated from robust Large Language Models (LLMs). Although these approaches deliver more performant models, they do not show sufficiently strong generalization ability as the training only relies on the provided demonstrations. In this paper, we propose the Self-refine Instruction-tuning method that elicits Smaller Language Models to self-refine their abilities. Our approach is based on a two-stage process, where reasoning abilities are first transferred between LLMs and Small Language Models (SLMs) via Instruction-tuning on demonstrations provided by LLMs, and then the instructed models Self-refine their abilities through preference optimization strategies. In particular, the second phase operates refinement heuristics based on the Direct Preference Optimization algorithm, where the SLMs are elicited to deliver a series of reasoning paths by automatically sampling the generated responses and providing rewards using ground truths from the LLMs. Results obtained on commonsense and math reasoning tasks show that this approach significantly outperforms Instruction-tuning in both in-domain and out-domain scenarios, aligning the reasoning abilities of Smaller and Larger Language Models.

Ritesh Karmaker, Shamoly Malaker

The article shows that W. B. Yeats' themes have drawn attention to how Ireland's history and culture have developed. Irish bard W. B. Yeats is a pioneer who is advancing his country's culture, society, and civilization. To maintain the political cacophony of his nation while maintaining the impression of Ireland's atmospheric layout in his literature, he is constantly seen upholding it. His philosophical thoughts and concepts soften the edge of Ireland's independence. He uses the themes to make a point about coming up with ideas. In his themes, he demonstrates his writing style and presentational structure. Yeats' perspective on politics systematizing literary themes has caught the enormity of writing impetus. This article shows a qualitative approach to the use of themes Yeats. The themes in W. B. Yeats' creation are on display, and the spontaneous attitudes that he used to shape the portrayal of Ireland are the presentation's heart and soul. The use of themes has been put to signify the motif of Yeats in Sherpur Sadar, Bangladesh. Astonishingly, the significance of Yeats influences the students living far away from Ireland. Here the concerned bodies opine on the various aspect of Yeats’ themes and a brief analysis of its’ connection with Japan’s Noh or Nogaku which essentially influenced Yeats’ works.

Sa'idatun Nishfullayli, Lea Santiar, Harni Kartika Ningsih

Making requests (irai) is a genre of spoken interaction that is taught from the basic level of learning Japanese as a foreign language. A request is one of the speech acts that may raise face-threatening potentials. Understanding the stages of request appropriate to Japanese culture is thus essential for Japanese learners to achieve successful conversation. Therefore, conversation pedagogy by using a discourse approach is essential. This study investigates a potential structure gap in Japanese making-requests conversations realized in actual settings and textbook conversational models. By employing genre theory and interpersonal discourse of “Negotiation” as a qualitative discourse analytic method from the Systemic Functional Linguistics (SFL) perspective, this paper describes the gaps and some factors that potentially influence the structure of Japanese making-request conversation. Data were obtained from conversational texts in the Japanese language corpus named Japanese Natural Conversation Corpus and Japanese textbooks for elementary and middle adult learners. Regarding the structure, the results show no difference between conversations in textbooks and authentic ones at the stage level, but both differ at the phase level. There is no introduction to the problem, additional explanation, and confirmation phases in textbook conversational models. In addition, the absence of the phases, the differences in pre-condition content between textbook and authentic conversations, also the length of the reasoning phase, are assumed to be influenced by relational status between participants (tenor) as well as the imposition degree of the requested object.

Sri Muryati, Bekti Setio Astuti

This study aims to determine the Japanese language skills needed by employees of PT. Java PT. Java Agritech Semarang. This research is the basis for Japanese language training activities that will be carried out as a form of cooperation between PT. Java Agritech Semarang and Japanese Language Study Program, so that training can run effectively and efficiently. Needs analysis is important to be carried out so that teachers can determine the right material, teaching methods, and references and can answer employee needs. The population of this study were employees of PT JAT Semarang not Japanese language professionals. Survey, questionnaire and interview methods are used to obtain data on employee needs for Japanese which refers to Target Needs and Learning Needs. . The results showed that 92.8% of respondents needed mastery of Japanese. Specifically, the mastery needed is work-related vocabulary, basic grammar and expressions in communicating. This aspect has an important role in four language skills, namely listening, reading, speaking, and writing. Of the four skills, respondents sequentially need speaking skills, listening skills, writing and reading. Speaking and listening skills include the ability to convey and identify greetings, name and point objects, ask and understand questions and simple explanations in Japanese. Reading and writing skills are only needed by 24% of respondents, to read and write emails in Japanese.

Vedant Palit, Rohan Pandey, Aryaman Arora et al.

Mechanistic interpretability seeks to understand the neural mechanisms that enable specific behaviors in Large Language Models (LLMs) by leveraging causality-based methods. While these approaches have identified neural circuits that copy spans of text, capture factual knowledge, and more, they remain unusable for multimodal models since adapting these tools to the vision-language domain requires considerable architectural changes. In this work, we adapt a unimodal causal tracing tool to BLIP to enable the study of the neural mechanisms underlying image-conditioned text generation. We demonstrate our approach on a visual question answering dataset, highlighting the causal relevance of later layer representations for all tokens. Furthermore, we release our BLIP causal tracing tool as open source to enable further experimentation in vision-language mechanistic interpretability by the community. Our code is available at https://github.com/vedantpalit/Towards-Vision-Language-Mechanistic-Interpretability.

Oluwatosin Ogundare, Subuola Sofolahan

This study investigates the potential of an ambulatory device that incorporates Large Language Models (LLMs) in cadence with other specialized ML models to assess anemia severity in sickle cell patients in real time. The device would rely on sensor data that measures angiogenic material levels to assess anemia severity, providing real-time information to patients and clinicians to reduce the frequency of vaso-occlusive crises because of the early detection of anemia severity, allowing for timely interventions and potentially reducing the likelihood of serious complications. The main challenges in developing such a device are the creation of a reliable non-invasive tool for angiogenic level assessment, a biophysics model and the practical consideration of an LLM communicating with emergency personnel on behalf of an incapacitated patient. A possible system is proposed, and the limitations of this approach are discussed.

Zaki Ainul Fadli, Tribuana Tunggadewi Kusumowardhani Baiquni

This study aims to determine the role of parents towards children in the short story 'Seibei to Hyotan' by Shiga Naoya published in 1913 in Japan. The research method used in this study is descriptive qualitative with data collection through literature studies. Data analysis is carried out by analyzing one of the elements of the short story structure, namely characterization and characterization, then understanding the relationship between the role of parents and children in this short story. The results showed that (1) the role of Seibei's parents towards Seibei was very important in shaping Seibei's character, (2) adults in this case the teacher from Seibei also played a role in the formation of Seibei's character, and (3) the impact of Seibei's parents' attitude towards Seibei was very large, one of which was Seibei's mistakes that recurred.

Barack W. Wanjawa, Lilian D. A. Wanzare, Florence Indede et al.

The need for Question Answering datasets in low resource languages is the motivation of this research, leading to the development of Kencorpus Swahili Question Answering Dataset, KenSwQuAD. This dataset is annotated from raw story texts of Swahili low resource language, which is a predominantly spoken in Eastern African and in other parts of the world. Question Answering (QA) datasets are important for machine comprehension of natural language for tasks such as internet search and dialog systems. Machine learning systems need training data such as the gold standard Question Answering set developed in this research. The research engaged annotators to formulate QA pairs from Swahili texts collected by the Kencorpus project, a Kenyan languages corpus. The project annotated 1,445 texts from the total 2,585 texts with at least 5 QA pairs each, resulting into a final dataset of 7,526 QA pairs. A quality assurance set of 12.5% of the annotated texts confirmed that the QA pairs were all correctly annotated. A proof of concept on applying the set to the QA task confirmed that the dataset can be usable for such tasks. KenSwQuAD has also contributed to resourcing of the Swahili language.

Martin Roth, Fabian Schäfer

An imagined conversation between Azuma Hiroki and an old man (maybe Jean-Jacques Rousseau) regarding their ideas about "Social Contract" and "General Will 2.0".

V. V. Shchepkin

The article traces how the image of the Ainu formed by Japaneseintellectuals in 18thand early 19thcenturies influenced the formation of Japan’spolicy towards this ethnic minority in the Meiji period (1868-1912).