nlp-low-resource-indo
Query: low resource Indonesian NLP
Results: 50
Date: 2026-07-07T18:52:41.711Z
1. NusaCrowd: A Call for Open and Reproducible NLP Research in Indonesian Languages
Authors: Samuel Cahyawijaya, Alham Fikri Aji, Holy Lovenia, Genta Indra Winata, Bryan Wilie, Rahmad Mahendra, Fajri Koto, David Moeljadi, Karissa Vincentio, Ade Romadhony, Ayu Purwarianti
Categories: cs.CL, cs.AI
Published: 2022-07-21
arXiv: 2207.10524v2
Abstract:
At the center of the underlying issues that halt Indonesian natural language processing (NLP) research advancement, we find data scarcity. Resources in Indonesian languages, especially the local ones, are extremely scarce and underrepresented. Many Indonesian researchers do not publish their dataset. Furthermore, the few public datasets that we have are scattered across different platforms, thus makes performing reproducible and data-centric research in Indonesian NLP even more arduous. Rising to this challenge, we initiate the first Indonesian NLP crowdsourcing effort, NusaCrowd. NusaCrowd strives to provide the largest datasheets aggregation with standardized data loading for NLP tasks in all Indonesian languages. By enabling open and centralized access to Indonesian NLP resources, we hope NusaCrowd can tackle the data scarcity problem hindering NLP progress in Indonesia and bring NLP practitioners to move towards collaboration.
2. A dissemination workshop for introducing young Italian students to NLP
Authors: Lucio Messina, Lucia Busso, Claudia Roberta Combei, Ludovica Pannitto, Alessio Miaschi, Gabriele Sarti, Malvina Nissim
Categories: cs.CL
Published: 2021-04-26
arXiv: 2104.12405v2
Abstract:
We describe and make available the game-based material developed for a laboratory run at several Italian science festivals to popularize NLP among young students.
3. Domain Adaptation of the Pyannote Diarization Pipeline for Conversational Indonesian Audio
Authors: Muhammad Daffa’i Rafi Prasetyo, Ramadhan Andika Putra, Zaidan Naufal Ilmi, Kurniawati Azizah
Categories: cs.SD
Published: 2026-01-07
arXiv: 2601.03684v1
Abstract:
This study presents a domain adaptation approach for speaker diarization targeting conversational Indonesian audio. We address the challenge of adapting an English-centric diarization pipeline to a low-resource language by employing synthetic data generation using neural Text-to-Speech technology. Experiments were conducted with varying training configurations, a small dataset (171 samples) and a large dataset containing 25 hours of synthetic speech. Results demonstrate that the baseline \texttt{pyannote/segmentation-3.0} model, trained on the AMI Corpus, achieves a Diarization Error Rate (DER) of 53.47% when applied zero-shot to Indonesian. Domain adaptation significantly improves performance, with the small dataset models reducing DER to 34.31% (1 epoch) and 34.81% (2 epochs). The model trained on the 25-hour dataset achieves the best performance with a DER of 29.24%, representing a 13.68% absolute improvement over the baseline while maintaining 99.06% Recall and 87.14% F1-Score.
4. Sejarah dan Perkembangan Teknik Natural Language Processing (NLP) Bahasa Indonesia: Tinjauan tentang sejarah, perkembangan teknologi, dan aplikasi NLP dalam bahasa Indonesia
Authors: Mukhlis Amien
Categories: cs.CL
Published: 2023-03-28
arXiv: 2304.02746v1
Abstract:
This study provides an overview of the history of the development of Natural Language Processing (NLP) in the context of the Indonesian language, with a focus on the basic technologies, methods, and practical applications that have been developed. This review covers developments in basic NLP technologies such as stemming, part-of-speech tagging, and related methods; practical applications in cross-language information retrieval systems, information extraction, and sentiment analysis; and methods and techniques used in Indonesian language NLP research, such as machine learning, statistics-based machine translation, and conflict-based approaches. This study also explores the application of NLP in Indonesian language industry and research and identifies challenges and opportunities in Indonesian language NLP research and development. Recommendations for future Indonesian language NLP research and development include developing more efficient methods and technologies, expanding NLP applications, increasing sustainability, further research into the potential of NLP, and promoting interdisciplinary collaboration. It is hoped that this review will help researchers, practitioners, and the government to understand the development of Indonesian language NLP and identify opportunities for further research and development.
5. Suryakala-Nusantara: Documenting Indonesian Sundials
Authors: Rhorom Priyatikanto
Categories: physics.pop-ph, astro-ph.IM
Published: 2013-12-10
arXiv: 1312.2742v1
Abstract:
Sundial is the ancient or classic timekeeper device, especially prior to the invention of mechanical clock. In the classical Islamic civilization, the daily movement of the Sun becomes main indicator of praying time, which can be deduced using sundial. This kind of device probably permeated to Indonesia during the Islamic acculturation. Since then, the development of astronomical knowledge, technology, art and architectural in classical Indonesia are partially reflected into sundial. These historical attractions of sundial demand comprehensive documentation and investigation of Indonesian sundial which are rarely found in the current literatures. The required spatial and temporal information regarding Indonesian sundial can be collected by general public through citizen science scheme. This concept may answer scientific curiosity of a research and also educate the people, expose them with science. In this article, general scheme of citizen science are discussed, its application for sundial study in Indonesia is proposed as Suryakala-Nusantara program.
6. NusaCrowd: Open Source Initiative for Indonesian NLP Resources
Authors: Samuel Cahyawijaya, Holy Lovenia, Alham Fikri Aji, Genta Indra Winata, Bryan Wilie, Rahmad Mahendra, Christian Wibisono, Ade Romadhony, Karissa Vincentio, Fajri Koto, Jennifer Santoso, David Moeljadi, Cahya Wirawan, Frederikus Hudi, Ivan Halim Parmonangan, Ika Alfina, Muhammad Satrio Wicaksono, Ilham Firdausi Putra, Samsul Rahmadani, Yulianti Oenang, Ali Akbar Septiandri, James Jaya, Kaustubh D. Dhole, Arie Ardiyanti Suryani, Rifki Afina Putri, Dan Su, Keith Stevens, Made Nindyatama Nityasya, Muhammad Farid Adilazuarda, Ryan Ignatius, Ryandito Diandaru, Tiezheng Yu, Vito Ghifari, Wenliang Dai, Yan Xu, Dyah Damapuspita, Cuk Tho, Ichwanul Muslim Karo Karo, Tirana Noor Fatyanosa, Ziwei Ji, Pascale Fung, Graham Neubig, Timothy Baldwin, Sebastian Ruder, Herry Sujaini, Sakriani Sakti, Ayu Purwarianti
Categories: cs.CL, cs.AI
Published: 2022-12-19
arXiv: 2212.09648v4
Abstract:
We present NusaCrowd, a collaborative initiative to collect and unify existing resources for Indonesian languages, including opening access to previously non-public resources. Through this initiative, we have brought together 137 datasets and 118 standardized data loaders. The quality of the datasets has been assessed manually and automatically, and their value is demonstrated through multiple experiments. NusaCrowd’s data collection enables the creation of the first zero-shot benchmarks for natural language understanding and generation in Indonesian and the local languages of Indonesia. Furthermore, NusaCrowd brings the creation of the first multilingual automatic speech recognition benchmark in Indonesian and the local languages of Indonesia. Our work strives to advance natural language processing (NLP) research for languages that are under-represented despite being widely spoken.
7. Teaching NLP with Bracelets and Restaurant Menus: An Interactive Workshop for Italian Students
Authors: Ludovica Pannitto, Lucia Busso, Claudia Roberta Combei, Lucio Messina, Alessio Miaschi, Gabriele Sarti, Malvina Nissim
Categories: cs.CL
Published: 2021-04-26
arXiv: 2104.12422v2
Abstract:
Although Natural Language Processing (NLP) is at the core of many tools young people use in their everyday life, high school curricula (in Italy) do not include any computational linguistics education. This lack of exposure makes the use of such tools less responsible than it could be and makes choosing computational linguistics as a university degree unlikely. To raise awareness, curiosity, and longer-term interest in young people, we have developed an interactive workshop designed to illustrate the basic principles of NLP and computational linguistics to high school Italian students aged between 13 and 18 years. The workshop takes the form of a game in which participants play the role of machines needing to solve some of the most common problems a computer faces in understanding language: from voice recognition to Markov chains to syntactic parsing. Participants are guided through the workshop with the help of instructors, who present the activities and explain core concepts from computational linguistics. The workshop was presented at numerous outlets in Italy between 2019 and 2021, both face-to-face and online.
8. Superstructures at low spin - high spin transitions
Authors: D. I. Khomskii, U. Löw
Categories: cond-mat.str-el
Published: 2001-06-07
arXiv: cond-mat/0106135v2
Abstract:
In many transition metal compounds, in particular those containing Fe^{2+} and Co^{3+}, there occur spin-state transitions between low-spin and high-spin (or intermediate-spin) states. We show that typical interactions between similar spin-state ions are short-range repulsion, and long-range interaction which can have different sign depending on the elastic anisotropy of the lattice and on the direction between respective ions. Due to such character of effective interactions at the spin-state transitions there may occur different superstructures – ordered arrangement of different spin-states, which in particular may have the form of stripes. The properties of the system TlSr_2CoO_5 for which such a superstructure was recently observed experimentally, are discussed from this point of view.
9. Location-based Twitter Filtering for the Creation of Low-Resource Language Datasets in Indonesian Local Languages
Authors: Mukhlis Amien, Chong Feng, Heyan Huang
Categories: cs.CL, cs.LG
Published: 2022-06-15
arXiv: 2206.07238v1
Abstract:
Twitter contains an abundance of linguistic data from the real world. We examine Twitter for user-generated content in low-resource languages such as local Indonesian. For NLP to work in Indonesian, it must consider local dialects, geographic context, and regional culture influence Indonesian languages. This paper identifies the problems we faced when constructing a Local Indonesian NLP dataset. Furthermore, we are developing a framework for creating, collecting, and classifying Local Indonesian datasets for NLP. Using twitter’s geolocation tool for automatic annotating.
10. IndoLEM and IndoBERT: A Benchmark Dataset and Pre-trained Language Model for Indonesian NLP
Authors: Fajri Koto, Afshin Rahimi, Jey Han Lau, Timothy Baldwin
Categories: cs.CL
Published: 2020-11-02
arXiv: 2011.00677v1
Abstract:
Although the Indonesian language is spoken by almost 200 million people and the 10th most spoken language in the world, it is under-represented in NLP research. Previous work on Indonesian has been hampered by a lack of annotated datasets, a sparsity of language resources, and a lack of resource standardization. In this work, we release the IndoLEM dataset comprising seven tasks for the Indonesian language, spanning morpho-syntax, semantics, and discourse. We additionally release IndoBERT, a new pre-trained language model for Indonesian, and evaluate it over IndoLEM, in addition to benchmarking it against existing resources. Our experiments show that IndoBERT achieves state-of-the-art performance over most of the tasks in IndoLEM.
11. Semi-Supervised Low-Resource Style Transfer of Indonesian Informal to Formal Language with Iterative Forward-Translation
Authors: Haryo Akbarianto Wibowo, Tatag Aziz Prawiro, Muhammad Ihsan, Alham Fikri Aji, Radityo Eko Prasojo, Rahmad Mahendra, Suci Fitriany
Categories: cs.CL
Published: 2020-11-06
arXiv: 2011.03286v2
Abstract:
In its daily use, the Indonesian language is riddled with informality, that is, deviations from the standard in terms of vocabulary, spelling, and word order. On the other hand, current available Indonesian NLP models are typically developed with the standard Indonesian in mind. In this work, we address a style-transfer from informal to formal Indonesian as a low-resource machine translation problem. We build a new dataset of parallel sentences of informal Indonesian and its formal counterpart. We benchmark several strategies to perform style transfer from informal to formal Indonesian. We also explore augmenting the training set with artificial forward-translated data. Since we are dealing with an extremely low-resource setting, we find that a phrase-based machine translation approach outperforms the Transformer-based approach. Alternatively, a pre-trained GPT-2 fined-tuned to this task performed equally well but costs more computational resource. Our findings show a promising step towards leveraging machine translation models for style transfer. Our code and data are available in https://github.com/haryoa/stif-indonesia
12. Domain-Specific Language Model Post-Training for Indonesian Financial NLP
Authors: Ni Putu Intan Maharani, Yoga Yustiawan, Fauzy Caesar Rochim, Ayu Purwarianti
Categories: cs.CL, cs.AI
Published: 2023-10-15
arXiv: 2310.09736v1
Abstract:
BERT and IndoBERT have achieved impressive performance in several NLP tasks. There has been several investigation on its adaption in specialized domains especially for English language. We focus on financial domain and Indonesian language, where we perform post-training on pre-trained IndoBERT for financial domain using a small scale of Indonesian financial corpus. In this paper, we construct an Indonesian self-supervised financial corpus, Indonesian financial sentiment analysis dataset, Indonesian financial topic classification dataset, and release a family of BERT models for financial NLP. We also evaluate the effectiveness of domain-specific post-training on sentiment analysis and topic classification tasks. Our findings indicate that the post-training increases the effectiveness of a language model when it is fine-tuned to domain-specific downstream tasks.
13. Comparative Analysis of AutoML and BiLSTM Models for Cyberbullying Detection on Indonesian Instagram Comments
Authors: Raihana Adelia Putri, Aisyah Musfirah, Anggi Puspita Ningrum, Luluk Muthoharoh, Ardika Satria, Martin Clinton Tosima Manullang
Categories: cs.CL
Published: 2026-04-29
arXiv: 2604.26229v1
Abstract:
This study compares machine learning and deep learning approaches for cyberbullying detection in Indonesian-language Instagram comments. Using a balanced dataset of 650 comments labeled as Bullying and Non-Bullying, the study evaluates Naive Bayes, Logistic Regression, and Support Vector Machine with TF-IDF features, as well as BiLSTM and BiLSTM with Bahdanau Attention. A preprocessing pipeline tailored to informal Indonesian text is applied, including slang normalization, stopword removal, and stemming. The results show that Logistic Regression performs best among the machine learning models, while BiLSTM with Attention achieves the strongest overall deep learning performance. The findings highlight the value of domain-specific preprocessing and show that although deep learning captures contextual patterns more effectively, machine learning remains a competitive option for resource-constrained deployments.
14. Field topologies in ideal and near ideal magnetohydrodynamics and vortex dynamics
Authors: B. C. Low
Categories: astro-ph.SR
Published: 2014-12-18
arXiv: 1412.6158v1
Abstract:
Magnetic field topology frozen in ideal magnetohydrodynamics (MHD) and its breakage in near ideal MHD are reviewed in two parts. The first part gives a physically complete description of the frozen in field topology, taking magnetic flux conservation as fundamental and treating four topics, Eulerian and Lagrangian descriptions of MHD, Chandrasekhar-Kendall and Euler-potential field representations, magnetic helicity, and inviscid vortex dynamics in comparison to ideal MHD. A corollary clarifies the challenge of achieving a high degree of the frozen in condition in numerical MHD. The second part treats field topology breakage centered on the Parker Magnetostatic Theorem on a general incompatibility of a continuous magnetic field with the dual demand of force free equilibrium and an arbitrarily prescribed, 3D field topology. Preserving field topology as a global constraint readily results in formation of tangential magnetic discontinuities, i.e., electric current sheets of zero thickness. A similar incompatibility is present in the steady, force and thermal balance of a heated radiating fluid subject to an anisotropic thermal flux conducted strictly along the frozen in magnetic field in the low beta limit. In a weakly resistive fluid the thinning of current sheets by these incompatibilities inevitably results in sheet dissipation, resistive heating and topological changes in the field despite the small resistivity. Faraday induction drives but also macroscopically limits this mode of energy dissipation, storing free energy in self organized, ideal MHD structures. This property of MHD turbulence captured by the Taylor hypothesis is reviewed in relation to the Sun’s corona, calling for a basic quantitative description of the breakdown of flux conservation in the low resistivity limit. A cylindrical, initial boundary value problem provides specificity in the review.
15. Bounds on the attractor dimension for magnetohydrodynamic channel flow with parallel magnetic field at low magnetic Reynolds number
Authors: Robert Low, Alban Potherat
Categories: physics.flu-dyn
Published: 2014-10-02
arXiv: 1410.0637v2
Abstract:
We investigate aspects of low-magnetic-Reynolds-number flow between two parallel, perfectly insulating walls, in the presence of an imposed magnetic field parallel to the bounding walls. We find a functional basis to describe the flow, well adapted to the problem of finding the attractor dimension, and which is also used in subsequent direct numerical simulation of these flows. For given Reynolds and Hartmann numbers, we obtain an upper bound for the dimension of the attractor by means of known bounds on the nonlinear inertial term and this functional basis for the flow. Three distinct flow regimes emerge: a quasi-isotropic 3D flow, a non-isotropic three-dimensional (3D) flow, and a 2D flow. We find the transition curves between these regimes in the space parameterized by Hartmann number Ha and attractor dimension $d_\text{att}$. We find how the attractor dimension scales as a function of Reynolds and Hartmann numbers (Re and Ha) in each regime. We also investigate the thickness of the boundary layer along the bounding wall, and find that in all regimes this scales as 1/Re, independently of the value of Ha, unlike Hartmann boundary layers found when the field is normal to the channel. The structure of the set of least dissipative modes is indeed quite different between these two cases but the properties of turbulence far from the walls (smallest scales and number of degrees of freedom) are found to be very similar.
16. Overview of the First Workshop on Language Models for Low-Resource Languages (LoResLM 2025)
Authors: Hansi Hettiarachchi, Tharindu Ranasinghe, Paul Rayson, Ruslan Mitkov, Mohamed Gaber, Damith Premasiri, Fiona Anting Tan, Lasitha Uyangodage
Categories: cs.CL, cs.AI
Published: 2024-12-20
arXiv: 2412.16365v1
Abstract:
The first Workshop on Language Models for Low-Resource Languages (LoResLM 2025) was held in conjunction with the 31st International Conference on Computational Linguistics (COLING 2025) in Abu Dhabi, United Arab Emirates. This workshop mainly aimed to provide a forum for researchers to share and discuss their ongoing work on language models (LMs) focusing on low-resource languages, following the recent advancements in neural language models and their linguistic biases towards high-resource languages. LoResLM 2025 attracted notable interest from the natural language processing (NLP) community, resulting in 35 accepted papers from 52 submissions. These contributions cover a broad range of low-resource languages from eight language families and 13 diverse research areas, paving the way for future possibilities and promoting linguistic inclusivity in NLP.
17. Constructing and Expanding Low-Resource and Underrepresented Parallel Datasets for Indonesian Local Languages
Authors: Joanito Agili Lopo, Radius Tanone
Categories: cs.CL
Published: 2024-04-01
arXiv: 2404.01009v1
Abstract:
In Indonesia, local languages play an integral role in the culture. However, the available Indonesian language resources still fall into the category of limited data in the Natural Language Processing (NLP) field. This is become problematic when build NLP model for these languages. To address this gap, we introduce Bhinneka Korpus, a multilingual parallel corpus featuring five Indonesian local languages. Our goal is to enhance access and utilization of these resources, extending their reach within the country. We explained in a detail the dataset collection process and associated challenges. Additionally, we experimented with translation task using the IBM Model 1 due to data constraints. The result showed that the performance of each language already shows good indications for further development. Challenges such as lexical variation, smoothing effects, and cross-linguistic variability are discussed. We intend to evaluate the corpus using advanced NLP techniques for low-resource languages, paving the way for multilingual translation models.
18. The Role of Stellar Feedback in the Chemical Evolution of a Low Mass Dwarf Galaxy
Authors: Andrew Emerick, Greg L. Bryan, Mordecai-Mark Mac Low
Categories: astro-ph.GA
Published: 2020-07-07
arXiv: 2007.03702v1
Abstract:
We investigate how each aspect of a multi-channel stellar feedback model drives the chemodynamical evolution of a low-mass, isolated dwarf galaxy using a suite of high-resolution simulations. Our model follows individual star particles sampled randomly from an adopted initial mass function, considering independently feedback from: supernovae; stellar radiation causing photoelectric heating of dust grains, ionization and associated heating, Lyman-Werner (LW) dissociation of H$_2$, and radiation pressure; and winds from massive main sequence (neglecting their energy input) and asymptotic giant branch (AGB) stars. Radiative transfer is done by ray tracing. We consider the effects each of these processes have on regulating the star formation rate, global properties, multi-phase interstellar medium (ISM), and driving of galactic winds. We follow individual metal species from distinct nucleosynthetic enrichment channels (AGB winds, massive star stellar winds, core collapse and Type Ia supernovae) and pay particular attention to how these feedback processes regulate metal mixing in the ISM, the metal content of outflows, and the stellar abundance patterns in our galaxy. We find that—for a low-metallicity, low-mass dwarf galaxy —stellar radiation, particularly ionizing radiation and LW radiation, are important sources of stellar feedback whose effects dominate over photoelectric heating and HI radiation pressure. However, feedback is coupled non-linearly, and the inclusion or exclusion of each process produces non-negligible effects. We find strong variations with: the star formation history; the ejection fractions of metals, mass, and energy; and the distribution of elements from different nucleosynthetic sources in both the gas and stars.
19. Deblur e-NeRF: NeRF from Motion-Blurred Events under High-speed or Low-light Conditions
Authors: Weng Fei Low, Gim Hee Lee
Categories: cs.CV, cs.GR, cs.RO, eess.SY
Published: 2024-09-26
arXiv: 2409.17988v1
Abstract:
The stark contrast in the design philosophy of an event camera makes it particularly ideal for operating under high-speed, high dynamic range and low-light conditions, where standard cameras underperform. Nonetheless, event cameras still suffer from some amount of motion blur, especially under these challenging conditions, in contrary to what most think. This is attributed to the limited bandwidth of the event sensor pixel, which is mostly proportional to the light intensity. Thus, to ensure that event cameras can truly excel in such conditions where it has an edge over standard cameras, it is crucial to account for event motion blur in downstream applications, especially reconstruction. However, none of the recent works on reconstructing Neural Radiance Fields (NeRFs) from events, nor event simulators, have considered the full effects of event motion blur. To this end, we propose, Deblur e-NeRF, a novel method to directly and effectively reconstruct blur-minimal NeRFs from motion-blurred events generated under high-speed motion or low-light conditions. The core component of this work is a physically-accurate pixel bandwidth model proposed to account for event motion blur under arbitrary speed and lighting conditions. We also introduce a novel threshold-normalized total variation loss to improve the regularization of large textureless patches. Experiments on real and novel realistically simulated sequences verify our effectiveness. Our code, event simulator and synthetic event dataset will be open-sourced.
20. Gas Loss by Ram Pressure Stripping and Internal Feedback From Low Mass Milky Way Satellites
Authors: Andrew Emerick, Mordecai-Mark Mac Low, Jana Grcevich, Andrea Gatto
Categories: astro-ph.GA
Published: 2016-05-09
arXiv: 1605.02746v1
Abstract:
The evolution of dwarf satellites of the Milky Way is affected by the combination of ram pressure and tidal stripping, and internal feedback from massive stars. We investigate gas loss processes in the smallest satellites of the Milky Way using three-dimensional, high resolution, idealized wind tunnel simulations, accounting for gas loss through both ram pressure stripping and expulsion by supernova feedback. Using initial conditions appropriate for a dwarf galaxy like Leo T, we investigate whether or not environmental gas stripping and internal feedback can quench these low mass galaxies on the expected timescales, shorter than 2 Gyr. We find that supernova feedback contributes negligibly to the stripping rate for these low star formation rate galaxies. However, we also find that ram pressure stripping is less efficient than expected in the stripping scenarios we consider. Our work suggests that, although ram pressure stripping can eventually completely strip these galaxies, other physics is likely at play to reconcile our computed stripping times with the rapid quenching timescales deduced from observations of low mass Milky Way dwarf galaxies. We discuss the roles additional physics may play in this scenario, including host-satellite tidal interactions, cored vs. cuspy dark matter profiles, reionization, and satellite pre-processing. We conclude that a proper accounting of these physics together is necessary to understand the quenching of low mass Milky Way satellites.
21. Simulating Metal Mixing of Both Common and Rare Enrichment Sources in a Low Mass Dwarf Galaxy
Authors: Andrew Emerick, Greg L. Bryan, Mordecai-Mark Mac Low
Categories: astro-ph.GA
Published: 2019-09-10
arXiv: 1909.04695v2
Abstract:
One-zone models constructed to match observed stellar abundance patterns have been used extensively to constrain the sites of nucleosynthesis with sophisticated libraries of stellar evolution and stellar yields. The metal mixing included in these models is usually highly simplified, although it is likely to be a significant driver of abundance evolution. In this work we use high-resolution hydrodynamics simulations to investigate how metals from individual enrichment events with varying source energies $E_{\rm ej}$ mix throughout the multi-phase interstellar medium (ISM) of a low-mass ($M_{\rm gas}=2\times 10^{6}$~M${\odot}$), low-metallicity, isolated dwarf galaxy. These events correspond to the characteristic energies of both common and exotic astrophysical sites of nucleosynthesis, including: asymptotic giant branch winds ($E{\rm ej}\sim$10$^{46}$~erg), neutron star-neutron star mergers ($E_{\rm ej}\sim$10$^{49}$~erg), supernovae ($E_{\rm ej}\sim$10$^{51}$
erg), and hypernovae ($E_{\rm ej}\sim$10$^{52}$Gyr), with a clear trend of increasing homogeneity for the more energetic events. Given these timescales, we conclude that the spatial distribution and frequency of events are important drivers of abundance homogeneity on large scales; rare, low $E_{\rm ej}$ events should be characterized by particularly broad abundance distributions. The source energy $E_{\rm ej}$ also correlates with the fraction of metals ejected in galactic winds, ranging anywhere from 60% at the lowest energy to 95% for hypernovae. We conclude by examining how the radial position, local ISM density, and global star formation rate influence these results.erg). We find the mixing timescales for individual enrichment sources in our dwarf galaxy to be long (100Myr–1
22. Towards Open-Ended Discovery for Low-Resource NLP
Authors: Bonaventure F. P. Dossou, Henri Aïdasso
Categories: cs.CL, cs.AI
Published: 2025-09-22
arXiv: 2510.01220v2
Abstract:
Natural Language Processing (NLP) for low-resource languages remains fundamentally constrained by the lack of textual corpora, standardized orthographies, and scalable annotation pipelines. While recent advances in large language models have improved cross-lingual transfer, they remain inaccessible to underrepresented communities due to their reliance on massive, pre-collected data and centralized infrastructure. In this position paper, we argue for a paradigm shift toward open-ended, interactive language discovery, where AI systems learn new languages dynamically through dialogue rather than static datasets. We contend that the future of language technology, particularly for low-resource and under-documented languages, must move beyond static data collection pipelines toward interactive, uncertainty-driven discovery, where learning emerges dynamically from human-machine collaboration instead of being limited to pre-existing datasets. We propose a framework grounded in joint human-machine uncertainty, combining epistemic uncertainty from the model with hesitation cues and confidence signals from human speakers to guide interaction, query selection, and memory retention. This paper is a call to action: we advocate a rethinking of how AI engages with human knowledge in under-documented languages, moving from extractive data collection toward participatory, co-adaptive learning processes that respect and empower communities while discovering and preserving the world’s linguistic diversity. This vision aligns with principles of human-centered AI, emphasizing interactive, cooperative model building between AI systems and speakers.
23. Benchmarking LightGBM and BiLSTM for Sentiment Analysis on Indonesian E-Commerce Reviews
Authors: Lidia Natasyah Marpaung, Vania Claresta, Iqfina Haula Halika, Luluk Muthoharoh, Ardika Satria, Martin Clinton Tosima Manullang
Categories: cs.CL
Published: 2026-05-02
arXiv: 2605.01322v1
Abstract:
This study presents a comparative analysis between two primary approaches in Natural Language Processing (NLP): Machine Learning (ML) utilizing the PyCaret AutoML framework, and Deep Learning (DL). The evaluation is conducted on a sentiment analysis task using an Indonesian e-commerce review dataset sourced from Hugging Face. The dataset, consisting of 15,000 samples, is partitioned into training, validation, and testing sets. The ML experiments compare LightGBM, Logistic Regression, and Support Vector Machine (SVM) algorithms, whereas the DL experiment implements a Bidirectional Long Short-Term Memory (BiLSTM) architecture. The experimental results demonstrate that the BiLSTM model outperforms all ML models, achieving an accuracy of 98.87% and an F1-Score of 98.87%. Meanwhile, LightGBM emerges as the best-performing ML model with an accuracy of 98.23% in a highly efficient training time. This research proves that the BiLSTM architecture is highly capable of capturing the sequential context of Indonesian review texts, making it the superior model for this specific classification task.
24. Baseline Systems For The 2025 Low-Resource Audio Codec Challenge
Authors: Yusuf Ziya Isik, Rafał Łaganowski
Categories: cs.SD, cs.LG
Published: 2025-09-30
arXiv: 2510.00264v3
Abstract:
The Low-Resource Audio Codec (LRAC) Challenge aims to advance neural audio coding for deployment in resource-constrained environments. The first edition focuses on low-resource neural speech codecs that must operate reliably under everyday noise and reverberation, while satisfying strict constraints on computational complexity, latency, and bitrate. Track 1 targets transparency codecs, which aim to preserve the perceptual transparency of input speech under mild noise and reverberation. Track 2 addresses enhancement codecs, which combine coding and compression with denoising and dereverberation. This paper presents the official baseline systems for both tracks in the 2025 LRAC Challenge. The baselines are convolutional neural codec models with Residual Vector Quantization, trained end-to-end using a combination of adversarial and reconstruction objectives. We detail the data filtering and augmentation strategies, model architectures, optimization procedures, and checkpoint selection criteria.
25. Parallel Gaussian Process Regression for Big Data: Low-Rank Representation Meets Markov Approximation
Authors: Kian Hsiang Low, Jiangbo Yu, Jie Chen, Patrick Jaillet
Categories: stat.ML, cs.DC, cs.LG
Published: 2014-11-17
arXiv: 1411.4510v1
Abstract:
The expressive power of a Gaussian process (GP) model comes at a cost of poor scalability in the data size. To improve its scalability, this paper presents a low-rank-cum-Markov approximation (LMA) of the GP model that is novel in leveraging the dual computational advantages stemming from complementing a low-rank approximate representation of the full-rank GP based on a support set of inputs with a Markov approximation of the resulting residual process; the latter approximation is guaranteed to be closest in the Kullback-Leibler distance criterion subject to some constraint and is considerably more refined than that of existing sparse GP models utilizing low-rank representations due to its more relaxed conditional independence assumption (especially with larger data). As a result, our LMA method can trade off between the size of the support set and the order of the Markov property to (a) incur lower computational cost than such sparse GP models while achieving predictive performance comparable to them and (b) accurately represent features/patterns of any scale. Interestingly, varying the Markov order produces a spectrum of LMAs with PIC approximation and full-rank GP at the two extremes. An advantage of our LMA method is that it is amenable to parallelization on multiple machines/cores, thereby gaining greater scalability. Empirical evaluation on three real-world datasets in clusters of up to 32 computing nodes shows that our centralized and parallel LMA methods are significantly more time-efficient and scalable than state-of-the-art sparse and full-rank GP regression methods while achieving comparable predictive performances.
26. Training a Neural Network in a Low-Resource Setting on Automatically Annotated Noisy Data
Authors: Michael A. Hedderich, Dietrich Klakow
Categories: cs.LG, cs.CL, stat.ML
Published: 2018-07-02
arXiv: 1807.00745v2
Abstract:
Manually labeled corpora are expensive to create and often not available for low-resource languages or domains. Automatic labeling approaches are an alternative way to obtain labeled data in a quicker and cheaper way. However, these labels often contain more errors which can deteriorate a classifier’s performance when trained on this data. We propose a noise layer that is added to a neural network architecture. This allows modeling the noise and train on a combination of clean and noisy data. We show that in a low-resource NER task we can improve performance by up to 35% by using additional, noisy data and handling the noise.
27. Stellar Radiation is Critical for Regulating Star Formation and Driving Outflows in Low Mass Dwarf Galaxies
Authors: Andrew Emerick, Greg L. Bryan, Mordecai-Mark Mac Low
Categories: astro-ph.GA
Published: 2018-08-01
arXiv: 1808.00468v2
Abstract:
Effective stellar feedback is used in models of galaxy formation to drive realistic galaxy evolution. Models typically include energy injection from supernovae as the dominant form of stellar feedback, often in some form of sub-grid recipe. However, it has been recently suggested that pre-SN feedback (stellar winds or radiation) is necessary in high-resolution simulations of galaxy evolution to properly regulate star formation and properties of the interstellar medium (ISM). Following these processes is computationally challenging, so many prescriptions model this feedback approximately, accounting for the local destruction of dense gas clouds around newly formed stars in lieu of a full radiative transfer calculation. In this work we examine high resolution simulations (1.8~pc) of an isolated dwarf galaxy with detailed stellar feedback tracked on a star-by-star basis. By following stellar ionizing radiation with an adaptive ray-tracing radiative transfer method, we test its importance in regulating star formation and driving outflows in this galaxy. We find that including ionizing radiation reduces the star formation rate (SFR) by over a factor of 5, and is necessary to produce the ISM conditions needed for supernovae to drive significant outflows. We find that a localized approximation for radiation feedback is sufficient to regulate the SFR on short timescales, but does not allow significant outflows. Short and long range radiation effects are both important in driving the evolution of our low-metallicity, low-mass dwarf galaxy. Generalizing these results to more massive galaxies would be a valuable avenue of future research.
28. Orbital migration of low-mass planets in evolutionary radiative models: Avoiding catastrophic infall
Authors: W. Lyra, S. -J. Paardekooper, M. -M. Mac Low
Categories: astro-ph.EP
Published: 2010-03-04
arXiv: 1003.0925v2
Abstract:
Outward migration of low-mass planets has recently been shown to be a possibility in non-barotropic disks. We examine the consequences of this result in evolutionary models of protoplanetary disks. Planet migration occurs towards equilibrium radii with zero torque. These radii themselves migrate inwards because of viscous accretion and photoevaporation. We show that as the surface density and temperature fall, the planet orbital migration and disk depletion timescales eventually become comparable, with the precise timing depending on the mass of the planet. When this occurs, the planet decouples from the equilibrium radius. At this time, however, the gas surface density is already too low to drive substantial further migration. A higher mass planet, of 10 Earth masses, can open a gap during the late evolution of the disk, and stops migrating. Low mass planets, with 1 or 0.1 Earth masses, released beyond 1 AU in our models, avoid migrating into the star. Our results provide support for the reduced migration rates adopted in recent planet population synthesis models.
29. Benchmarking PyCaret AutoML Against IndoBERT Fine-Tuning for Sentiment Analysis on Indonesian IKN Twitter Data
Authors: Mutia Alfi Mayzaroh, Dwi Fitria Ningsih, Nindi Destriani, Martin C. T. Manullang
Categories: cs.CL
Published: 2026-04-28
arXiv: 2604.25392v1
Abstract:
This paper benchmarks a classical machine learning approach based on PyCaret AutoML against a deep learning approach based on IndoBERT fine-tuning for binary sentiment analysis of Indonesian-language Twitter comments related to Ibu Kota Nusantara (IKN). The dataset contains 1,472 manually labeled samples, consisting of 780 negative and 692 positive comments. In the machine learning setting, Logistic Regression, Naive Bayes, and Support Vector Machine were evaluated using 10-fold cross-validation, with Logistic Regression achieving the best performance among the classical models at 77.57% accuracy and 77.17% F1-score. In the deep learning setting, the indobenchmark/indobert-base-p1 model was fine-tuned for five epochs and achieved 89.59% test accuracy and 89.37% F1-score. The results show that IndoBERT substantially outperforms the machine learning baselines, highlighting the effectiveness of Transformer-based contextual representations for informal Indonesian social media text.
30. Improving the FAIRness and Sustainability of the NHGRI Resources Ecosystem
Authors: Larry Babb, Carol Bult, Vincent J. Carey, Robert J. Carroll, Benjamin C. Hitz, Chris J. Mungall, Heidi L. Rehm, Michael C. Schatz, Alex Wagner, NHGRI Resource Workshop Community
Categories: q-bio.GN
Published: 2025-08-19
arXiv: 2508.13498v1
Abstract:
In 2024, NHGRI-funded genomic resource projects completed a Self-Assessment Tool (SAT) and interviews to evaluate their application of FAIR (Findable, Accessible, Interoperable, Reusable) principles and sustainability. Key challenges were identified in metadata tools, data curation, variant identifiers, and data processing. Addressing these needs, we engaged the community through webinars and discussions, leading to a two-day workshop in March 2025. The workshop developed targeted recommendations, including improving transparency, standardizing identifiers, enhancing usability, implementing APIs, leveraging AI/ML for curation, and evaluating impact. These outcomes provide a framework for advancing FAIR practices, fostering collaboration, and strengthening the sustainability of NHGRI resources.
31. Parallel Gaussian Process Regression with Low-Rank Covariance Matrix Approximations
Authors: Jie Chen, Nannan Cao, Kian Hsiang Low, Ruofei Ouyang, Colin Keng-Yan Tan, Patrick Jaillet
Categories: stat.ML, cs.DC, cs.LG
Published: 2013-05-24
arXiv: 1305.5826v1
Abstract:
Gaussian processes (GP) are Bayesian non-parametric models that are widely used for probabilistic regression. Unfortunately, it cannot scale well with large data nor perform real-time predictions due to its cubic time cost in the data size. This paper presents two parallel GP regression methods that exploit low-rank covariance matrix approximations for distributing the computational load among parallel machines to achieve time efficiency and scalability. We theoretically guarantee the predictive performances of our proposed parallel GPs to be equivalent to that of some centralized approximate GP regression methods: The computation of their centralized counterparts can be distributed among parallel machines, hence achieving greater time efficiency and scalability. We analytically compare the properties of our parallel GPs such as time, space, and communication complexity. Empirical evaluation on two real-world datasets in a cluster of 20 computing nodes shows that our parallel GPs are significantly more time-efficient and scalable than their centralized counterparts and exact/full GP while achieving predictive performances comparable to full GP.
32. IVOA Recommendation: Resource Metadata for the Virtual Observatory Version 1.12
Authors: Robert Hanisch, the IVOA Resource Registry Working Group, the NVO Metadata Working Group
Categories: astro-ph.IM
Published: 2011-10-03
arXiv: 1110.0514v1
Abstract:
An essential capability of the Virtual Observatory is a means for describing what data and computational facilities are available where, and once identified, how to use them. The data themselves have associated metadata (e.g., FITS keywords), and similarly we require metadata about data collections and data services so that VO users can easily find information of interest. Furthermore, such metadata are needed in order to manage distributed queries efficiently; if a user is interested in finding x-ray images there is no point in querying the HST archive, for example. In this document we suggest an architecture for resource and service metadata and describe the relationship of this architecture to emerging Web Services standards. We also define an initial set of metadata concepts.
33. Improving Multilingual Neural Machine Translation For Low-Resource Languages: French,English - Vietnamese
Authors: Thi-Vinh Ngo, Phuong-Thai Nguyen, Thanh-Le Ha, Khac-Quy Dinh, Le-Minh Nguyen
Categories: cs.CL, cs.LG
Published: 2020-12-16
arXiv: 2012.08743v2
Abstract:
Prior works have demonstrated that a low-resource language pair can benefit from multilingual machine translation (MT) systems, which rely on many language pairs’ joint training. This paper proposes two simple strategies to address the rare word issue in multilingual MT systems for two low-resource language pairs: French-Vietnamese and English-Vietnamese. The first strategy is about dynamical learning word similarity of tokens in the shared space among source languages while another one attempts to augment the translation ability of rare words through updating their embeddings during the training. Besides, we leverage monolingual data for multilingual MT systems to increase the amount of synthetic parallel corpora while dealing with the data sparsity problem. We have shown significant improvements of up to +1.62 and +2.54 BLEU points over the bilingual baseline systems for both language pairs and released our datasets for the research community.
34. The effect of feedback and reionization on star formation in low-mass dwarf galaxy haloes
Authors: Christine M. Simpson, Greg L. Bryan, Kathryn V. Johnston, Britton D. Smith, Mordecai-Mark Mac Low, Sanjib Sharma, Jason Tumlinson
Categories: astro-ph.GA, astro-ph.CO
Published: 2012-11-05
arXiv: 1211.1071v2
Abstract:
We simulate the evolution of a 10^9 Msun dark matter halo in a cosmological setting with an adaptive-mesh refinement code as an analogue to local low luminosity dwarf irregular and dwarf spheroidal galaxies. The primary goal of our study is to investigate the roles of reionization and supernova feedback in determining the star formation histories of low mass dwarf galaxies. We include a wide range of physical effects, including metal cooling, molecular hydrogen formation and cooling, photoionization and photodissociation from a metagalactic background, a simple prescription for self-shielding, star formation, and a simple model for supernova driven energetic feedback. We carry out simulations excluding each major effect in turn. We find that reionization is primarily responsible for expelling most of the gas in our simulations, but that supernova feedback is required to disperse the dense, cold gas in the core of the halo. Moreover, we show that the timing of reionization can produce an order of magnitude difference in the final stellar mass of the system. For our full physics run with reionization at z=9, we find a stellar mass of about 10^5 Msun at z=0, and a mass-to-light ratio within the half-light radius of approximately 130 Msun/Lsun, consistent with observed low-luminosity dwarfs. However, the resulting median stellar metallicity is 0.06 Zsun, considerably larger than observed systems. In addition, we find star formation is truncated between redshifts 4 and 7, at odds with the observed late time star formation in isolated dwarf systems but in agreement with Milky Way ultrafaint dwarf spheroidals. We investigate the efficacy of energetic feedback in our simple thermal-energy driven feedback scheme, and suggest that it may still suffer from excessive radiative losses, despite reaching stellar particle masses of about 100 Msun, and a comoving spatial resolution of 11 pc.
35. Star formation at very low metallicity. II: On the insignificance of metal-line cooling during the early stages of gravitational collapse
Authors: A. -K. Jappsen, S. C. O. Glover, R. S. Klessen, M. -M. Mac Low
Categories: astro-ph
Published: 2005-11-14
arXiv: astro-ph/0511400v3
Abstract:
We study the influence of low levels of metal enrichment on the cooling and collapse of ionized gas in small protogalactic halos using three-dimensional, smoothed particle hydrodynamics simulations. Our initial conditions represent protogalaxies forming within a fossil HII region – a previously ionized HII region which has not yet had time to cool and recombine. Prior to cosmological reionization, such regions should be relatively common, since the characteristic lifetimes of the likely ionizing sources are significantly shorter than a Hubble time. We show that in these regions, H_2 is the dominant and most effective coolant, and that it is the amount of H_2 formed that determines whether or not the gas can collapse and form stars. At the low metallicities (Z < 10^{-3} Z_sun) thought to be associated with the transition from population III to early population II star formation, metal line cooling has an almost negligible effect on the evolution of low density gas, altering the density and temperature evolution of the gas by less than 1% compared to the metal-free case at densities below 1 cm^{-3} and temperatures above 2000 K. Although there is evidence that metal line cooling becomes more effective at higher density, we find no significant differences in behaviour from the metal-free case at any density below our sink particle creation threshold at n = 500 cm^{-3}. Increasing the metallicity also increases the importance of metal line cooling, but it does not significantly affect the dynamical evolution of the low density gas until Z = 0.1 Z_sun. This result holds regardless of whether or not an ultraviolet background is present.
36. Speechless: Speech Instruction Training Without Speech for Low Resource Languages
Authors: Alan Dao, Dinh Bach Vu, Huy Hoang Ha, Tuan Le Duc Anh, Shreyas Gopal, Yue Heng Yeo, Warren Keng Hoong Low, Eng Siong Chng, Jia Qi Yip
Categories: eess.AS, cs.CL, cs.SD
Published: 2025-05-23
arXiv: 2505.17417v1
Abstract:
The rapid growth of voice assistants powered by large language models (LLM) has highlighted a need for speech instruction data to train these systems. Despite the abundance of speech recognition data, there is a notable scarcity of speech instruction data, which is essential for fine-tuning models to understand and execute spoken commands. Generating high-quality synthetic speech requires a good text-to-speech (TTS) model, which may not be available to low resource languages. Our novel approach addresses this challenge by halting synthesis at the semantic representation level, bypassing the need for TTS. We achieve this by aligning synthetic semantic representations with the pre-trained Whisper encoder, enabling an LLM to be fine-tuned on text instructions while maintaining the ability to understand spoken instructions during inference. This simplified training process is a promising approach to building voice assistant for low-resource languages.
37. Magnetic fields do not suppress global star formation in low metallicity dwarf galaxies
Authors: David J. Whitworth, Rowan J. Smith, Ralf S. Klessen, Mordecai-Mark Mac Low, Simon C. O. Glover, Robin Tress, Rudiger Pakmor, Juan D. Soler
Categories: astro-ph.GA
Published: 2022-10-10
arXiv: 2210.04922v2
Abstract:
Many studies concluded that magnetic fields suppress star formation in molecular clouds and Milky Way like galaxies. However, most of these studies are based on fully developed fields that have reached the saturation level, with little work on investigating how an initial weak primordial field affects star formation in low metallicity environments. In this paper, we investigate the impact of a weak initial field on low metallicity dwarf galaxies. We perform high-resolution AREPO simulations of five isolated dwarf galaxies. Two models are hydrodynamical, two start with a primordial magnetic field of 10$^{-6} μ$G and different sub-solar metallicities, and one starts with a saturated field of 10$^{-2} μ$G. All models include a non-equilibrium, time-dependent chemical network that includes the effects of gas shielding from the ambient ultraviolet field. Sink particles form directly from the gravitational collapse of gas and are treated as star-forming clumps that can accrete gas. We vary the ambient uniform far ultraviolet field, and cosmic ray ionization rate between 1% and 10% of solar values. We find that the magnetic field has little impact on the global star formation rate, which is in tension with some previously published results. We further find that the initial field strength has little impact on the global star formation rate. We show that an increase in the mass fractions of both molecular hydrogen and cold gas, along with changes in the perpendicular gas velocity dispersion and the magnetic field acting in the weak-field model, overcome the expected suppression in star formation.
38. GR-NLP-TOOLKIT: An Open-Source NLP Toolkit for Modern Greek
Authors: Lefteris Loukas, Nikolaos Smyrnioudis, Chrysa Dikonomaki, Spyros Barbakos, Anastasios Toumazatos, John Koutsikakis, Manolis Kyriakakis, Mary Georgiou, Stavros Vassos, John Pavlopoulos, Ion Androutsopoulos
Categories: cs.CL, cs.AI, cs.SE
Published: 2024-12-11
arXiv: 2412.08520v1
Abstract:
We present GR-NLP-TOOLKIT, an open-source natural language processing (NLP) toolkit developed specifically for modern Greek. The toolkit provides state-of-the-art performance in five core NLP tasks, namely part-of-speech tagging, morphological tagging, dependency parsing, named entity recognition, and Greeklishto-Greek transliteration. The toolkit is based on pre-trained Transformers, it is freely available, and can be easily installed in Python (pip install gr-nlp-toolkit). It is also accessible through a demonstration platform on HuggingFace, along with a publicly available API for non-commercial use. We discuss the functionality provided for each task, the underlying methods, experiments against comparable open-source toolkits, and future possible enhancements. The toolkit is available at: https://github.com/nlpaueb/gr-nlp-toolkit
39. Low-Resource Clickbait Spoiling for Indonesian via Question Answering
Authors: Ni Putu Intan Maharani, Ayu Purwarianti, Alham Fikri Aji
Categories: cs.CL, cs.AI
Published: 2023-10-12
arXiv: 2310.08085v1
Abstract:
Clickbait spoiling aims to generate a short text to satisfy the curiosity induced by a clickbait post. As it is a newly introduced task, the dataset is only available in English so far. Our contributions include the construction of manually labeled clickbait spoiling corpus in Indonesian and an evaluation on using cross-lingual zero-shot question answering-based models to tackle clikcbait spoiling for low-resource language like Indonesian. We utilize selection of multilingual language models. The experimental results suggest that XLM-RoBERTa (large) model outperforms other models for phrase and passage spoilers, meanwhile, mDeBERTa (base) model outperforms other models for multipart spoilers.
40. How Good Is NLP? A Sober Look at NLP Tasks through the Lens of Social Impact
Authors: Zhijing Jin, Geeticka Chauhan, Brian Tse, Mrinmaya Sachan, Rada Mihalcea
Categories: cs.CL, cs.AI, cs.CY, cs.LG
Published: 2021-06-04
arXiv: 2106.02359v3
Abstract:
Recent years have seen many breakthroughs in natural language processing (NLP), transitioning it from a mostly theoretical field to one with many real-world applications. Noting the rising number of applications of other machine learning and AI techniques with pervasive societal impact, we anticipate the rising importance of developing NLP technologies for social good. Inspired by theories in moral philosophy and global priorities research, we aim to promote a guideline for social good in the context of NLP. We lay the foundations via the moral philosophy definition of social good, propose a framework to evaluate the direct and indirect real-world impact of NLP tasks, and adopt the methodology of global priorities research to identify priority causes for NLP research. Finally, we use our theoretical framework to provide some practical guidelines for future NLP research for social good. Our data and code are available at http://github.com/zhijing-jin/nlp4sg_acl2021. In addition, we curate a list of papers and resources on NLP for social good at https://github.com/zhijing-jin/NLP4SocialGood_Papers.
41. IndoNLU: Benchmark and Resources for Evaluating Indonesian Natural Language Understanding
Authors: Bryan Wilie, Karissa Vincentio, Genta Indra Winata, Samuel Cahyawijaya, Xiaohong Li, Zhi Yuan Lim, Sidik Soleman, Rahmad Mahendra, Pascale Fung, Syafri Bahar, Ayu Purwarianti
Categories: cs.CL
Published: 2020-09-11
arXiv: 2009.05387v3
Abstract:
Although Indonesian is known to be the fourth most frequently used language over the internet, the research progress on this language in the natural language processing (NLP) is slow-moving due to a lack of available resources. In response, we introduce the first-ever vast resource for the training, evaluating, and benchmarking on Indonesian natural language understanding (IndoNLU) tasks. IndoNLU includes twelve tasks, ranging from single sentence classification to pair-sentences sequence labeling with different levels of complexity. The datasets for the tasks lie in different domains and styles to ensure task diversity. We also provide a set of Indonesian pre-trained models (IndoBERT) trained from a large and clean Indonesian dataset Indo4B collected from publicly available sources such as social media texts, blogs, news, and websites. We release baseline models for all twelve tasks, as well as the framework for benchmark evaluation, and thus it enables everyone to benchmark their system performances.
42. TajPersLexon: A Tajik-Persian Lexical Resource and Hybrid Model for Cross-Script Low-Resource NLP
Authors: Mullosharaf K. Arabov
Categories: cs.CL
Published: 2026-05-07
arXiv: 2605.06886v1
Abstract:
This work introduces TajPersLexon, a curated Tajik–Persian parallel lexical resource of 40,112 word and short-phrase pairs for cross-script lexical retrieval, transliteration, and alignment in low-resource settings. We conduct a comprehensive CPU-only benchmark comparing three methodological families: (i) a lightweight hybrid pipeline, (ii) neural sequence-to-sequence models, and (iii) retrieval methods. Our evaluation establishes that the task is essentially solvable, with neural and retrieval baselines achieving 98-99% top-1 accuracy. Crucially, we demonstrate that while large multilingual sentence transformers fail on this exact lexical matching, our interpretable hybrid model offers a favorable accuracy-efficiency trade-off for practical applications, achieving 96.4% accuracy in an OCR post-correction task. All experiments use fixed random seeds for full reproducibility. The dataset, code, and models will be publicly released.
43. Building Dialogue Understanding Models for Low-resource Language Indonesian from Scratch
Authors: Donglin Di, Weinan Zhang, Yue Zhang, Fanglin Wang
Categories: cs.CL
Published: 2024-10-24
arXiv: 2410.18430v1
Abstract:
Making use of off-the-shelf resources of resource-rich languages to transfer knowledge for low-resource languages raises much attention recently. The requirements of enabling the model to reach the reliable performance lack well guided, such as the scale of required annotated data or the effective framework. To investigate the first question, we empirically investigate the cost-effectiveness of several methods to train the intent classification and slot-filling models for Indonesia (ID) from scratch by utilizing the English data. Confronting the second challenge, we propose a Bi-Confidence-Frequency Cross-Lingual transfer framework (BiCF), composed by
BiCF Mixing'',Latent Space Refinement’’ and ``Joint Decoder’’, respectively, to tackle the obstacle of lacking low-resource language dialogue data. Extensive experiments demonstrate our framework performs reliably and cost-efficiently on different scales of manually annotated Indonesian data. We release a large-scale fine-labeled dialogue dataset (ID-WOZ) and ID-BERT of Indonesian for further research.
44. Characterizing the Efficiency vs. Accuracy Trade-off for Long-Context NLP Models
Authors: Phyllis Ang, Bhuwan Dhingra, Lisa Wu Wills
Categories: cs.CL, cs.LG
Published: 2022-04-15
arXiv: 2204.07288v1
Abstract:
With many real-world applications of Natural Language Processing (NLP) comprising of long texts, there has been a rise in NLP benchmarks that measure the accuracy of models that can handle longer input sequences. However, these benchmarks do not consider the trade-offs between accuracy, speed, and power consumption as input sizes or model sizes are varied. In this work, we perform a systematic study of this accuracy vs. efficiency trade-off on two widely used long-sequence models - Longformer-Encoder-Decoder (LED) and Big Bird - during fine-tuning and inference on four datasets from the SCROLLS benchmark. To study how this trade-off differs across hyperparameter settings, we compare the models across four sequence lengths (1024, 2048, 3072, 4096) and two model sizes (base and large) under a fixed resource budget. We find that LED consistently achieves better accuracy at lower energy costs than Big Bird. For summarization, we find that increasing model size is more energy efficient than increasing sequence length for higher accuracy. However, this comes at the cost of a large drop in inference speed. For question answering, we find that smaller models are both more efficient and more accurate due to the larger training batch sizes possible under a fixed resource budget.
45. Hybrid TF–IDF Logistic Regression and MLP Neural Baseline for Indonesian Three-Class Sentiment Analysis on Social Media Text
Authors: Allya Nurul Islami Pasha, Eka Fidiya Putri, Luluk Muthoharoh, Ardika Satria, Martin C. T. Manullang
Categories: cs.CL
Published: 2026-05-08
arXiv: 2605.07793v1
Abstract:
This paper presents a compact three-class sentiment analysis study for Indonesian social media text. The task is formulated with positive, negative, and neutral outputs derived from a fine-grained emotion dataset. The proposed practical baseline combines TF–IDF text features, three lightweight numeric metadata features, and a balanced multinomial Logistic Regression classifier. For comparison, the study also includes a neural baseline using a two-layer multilayer perceptron (MLP) over the same hybrid feature representation. The dataset originally contains 732 rows and 191 fine-grained emotion labels; after cleaning, deduplication, and label remapping, 707 samples remain with an imbalanced distribution of 459 positive, 188 negative, and 60 neutral instances. Experimental results show that the Logistic Regression deployment model reaches 0.8028 accuracy, 0.8003 weighted F1, and 0.7276 macro F1, while project documentation reports a higher-accuracy but non-production MLP baseline. These findings indicate that careful preprocessing, interpretable feature engineering, and class balancing remain competitive for small Indonesian sentiment datasets, whereas the neural baseline is better treated as a comparative experiment than as the default deployment model.
46. NLP-ADBench: NLP Anomaly Detection Benchmark
Authors: Yuangang Li, Jiaqi Li, Zhuo Xiao, Tiankai Yang, Yi Nian, Xiyang Hu, Yue Zhao
Categories: cs.CL, cs.LG
Published: 2024-12-06
arXiv: 2412.04784v2
Abstract:
Anomaly detection (AD) is an important machine learning task with applications in fraud detection, content moderation, and user behavior analysis. However, AD is relatively understudied in a natural language processing (NLP) context, limiting its effectiveness in detecting harmful content, phishing attempts, and spam reviews. We introduce NLP-ADBench, the most comprehensive NLP anomaly detection (NLP-AD) benchmark to date, which includes eight curated datasets and 19 state-of-the-art algorithms. These span 3 end-to-end methods and 16 two-step approaches that adapt classical, non-AD methods to language embeddings from BERT and OpenAI. Our empirical results show that no single model dominates across all datasets, indicating a need for automated model selection. Moreover, two-step methods with transformer-based embeddings consistently outperform specialized end-to-end approaches, with OpenAI embeddings outperforming those of BERT. We release NLP-ADBench at https://github.com/USC-FORTIS/NLP-ADBench, providing a unified framework for NLP-AD and supporting future investigations.
47. MyCulture: Exploring Malaysia’s Diverse Culture under Low-Resource Language Constraints
Authors: Zhong Ken Hew, Jia Xin Low, Sze Jue Yang, Chee Seng Chan
Categories: cs.CL, cs.AI
Published: 2025-08-07
arXiv: 2508.05429v2
Abstract:
Large Language Models (LLMs) often exhibit cultural biases due to training data dominated by high-resource languages like English and Chinese. This poses challenges for accurately representing and evaluating diverse cultural contexts, particularly in low-resource language settings. To address this, we introduce MyCulture, a benchmark designed to comprehensively evaluate LLMs on Malaysian culture across six pillars: arts, attire, customs, entertainment, food, and religion presented in Bahasa Melayu. Unlike conventional benchmarks, MyCulture employs a novel open-ended multiple-choice question format without predefined options, thereby reducing guessing and mitigating format bias. We provide a theoretical justification for the effectiveness of this open-ended structure in improving both fairness and discriminative power. Furthermore, we analyze structural bias by comparing model performance on structured versus free-form outputs, and assess language bias through multilingual prompt variations. Our evaluation across a range of regional and international LLMs reveals significant disparities in cultural comprehension, highlighting the urgent need for culturally grounded and linguistically inclusive benchmarks in the development and assessment of LLMs.
48. Star Formation at Very Low Metallicity. V. The Greater Importance of Initial Conditions Compared to Metallicity Thresholds
Authors: A. -K. Jappsen, M. -M. Mac Low, S. C. O. Glover, R. S. Klessen, S. Kitsionas
Categories: astro-ph
Published: 2008-10-10
arXiv: 0810.1867v2
Abstract:
The formation of the first stars out of metal-free gas appears to result in stars at least an order of magnitude more massive than in the present-day case. We here consider what controls the transition from a primordial to a modern initial mass function. It has been proposed that this occurs when effective metal line cooling occurs at a metallicity threshold of Z/Z_sun > 10^{-3.5}. We study the influence of low levels of metal enrichment on the cooling and collapse of initially ionized gas in small protogalactic halos using three-dimensional, smoothed particle hydrodynamics simulations with particle splitting. Our initial conditions represent protogalaxies forming within a previously ionized H ii region that has not yet had time to cool and recombine. These differ considerably from those used in simulations predicting a metallicity threshold, where the gas was initially cold and only partially ionized. In the centrally condensed potential that we study here, a wide variety of initial conditions for the gas yield a monolithic central collapse. Our models show no fragmentation during collapse to number densities as high as 10^5 cm^{-3}, for metallicities reaching as high as 10^{-1} Z_sun in one rotating case, far above the threshold suggested by previous work. Rotation allows for the formation of gravitationally stable gas disks over large fractions of the local Hubble time. Turbulence slows the growth of the central density slightly, but both spherically symmetric and turbulent initial conditions collapse and form a single sink particle. We therefore argue that fragmentation at moderate density depends on the initial conditions for star formation more than on the metal abundances present.
49. SPRING Lab IITM’s submission to Low Resource Indic Language Translation Shared Task
Authors: Hamees Sayed, Advait Joglekar, Srinivasan Umesh
Categories: cs.CL, cs.AI
Published: 2024-11-01
arXiv: 2411.00727v2
Abstract:
We develop a robust translation model for four low-resource Indic languages: Khasi, Mizo, Manipuri, and Assamese. Our approach includes a comprehensive pipeline from data collection and preprocessing to training and evaluation, leveraging data from WMT task datasets, BPCC, PMIndia, and OpenLanguageData. To address the scarcity of bilingual data, we use back-translation techniques on monolingual datasets for Mizo and Khasi, significantly expanding our training corpus. We fine-tune the pre-trained NLLB 3.3B model for Assamese, Mizo, and Manipuri, achieving improved performance over the baseline. For Khasi, which is not supported by the NLLB model, we introduce special tokens and train the model on our Khasi corpus. Our training involves masked language modelling, followed by fine-tuning for English-to-Indic and Indic-to-English translations.
50. Teaching NLP outside Linguistics and Computer Science classrooms: Some challenges and some opportunities
Authors: Sowmya Vajjala
Categories: cs.CL
Published: 2021-05-03
arXiv: 2105.00895v1
Abstract:
NLP’s sphere of influence went much beyond computer science research and the development of software applications in the past decade. We see people using NLP methods in a range of academic disciplines from Asian Studies to Clinical Oncology. We also notice the presence of NLP as a module in most of the data science curricula within and outside of regular university setups. These courses are taken by students from very diverse backgrounds. This paper takes a closer look at some issues related to teaching NLP to these diverse audiences based on my classroom experiences, and identifies some challenges the instructors face, particularly when there is no ecosystem of related courses for the students. In this process, it also identifies a few challenge areas for both NLP researchers and tool developers.