Hieu Tran

Hugo Laurençon, Lucile Saulnier, Thomas Wang et al. · huggingface

As language models grow ever larger, the need for large-scale high-quality text datasets has never been more pressing, especially in multilingual settings. The BigScience workshop, a 1-year international and multidisciplinary initiative, was formed with the goal of researching and training large language models as a values-driven undertaking, putting issues of ethics, harm, and governance in the foreground. This paper documents the data creation and curation efforts undertaken by BigScience to assemble the Responsible Open-science Open-collaboration Text Sources (ROOTS) corpus, a 1.6TB dataset spanning 59 languages that was used to train the 176-billion-parameter BigScience Large Open-science Open-access Multilingual (BLOOM) language model. We further release a large initial subset of the corpus and analyses thereof, and hope to empower large-scale monolingual and multilingual modeling projects with both the data and the processing tools, as well as stimulate research around this large multilingual corpus.

Kelechi G. Kalu, Hieu Tran, Santiago Torres-Arias et al.

Identity-based software signing tools aim to make software artifact provenance verifiable while reducing the operational burden of long-lived key management. However, there is limited cross-tool longitudinal evidence about which usability problems arise in practice and how those problems evolve as tools mature. This gap matters because unusable signing and verification workflows can lead to incomplete adoption, misconfiguration, or skipped verification, undermining intended integrity guarantees. We conducted the first mining-software-repositories study of five open-source identity-based signing ecosystems: Sigstore, OpenPubKey, HashiCorp Vault, Keyfactor, and Notary v2. We analyzed approximately 3,900 GitHub issues from Nov. 2021 to Nov. 2025. We coded each issue for the reported usability concern and the implicated architectural component, and compared patterns across tools and over time. Across ecosystems, reported concerns concentrate in verification workflows, policy and configuration surfaces, and integration boundaries. Longitudinal Poisson trend analysis shows substantial declines in reported issues for most ecosystems. However, across usability themes, workflow- and documentation-related concerns decline unevenly across tools and concern types, and verification workflows and configuration surfaces remain persistent friction points. These results indicate that identity-based signing reduces some usability burdens while relocating complexity to verification semantics, policy configuration, and deployment integration. Designing future signing ecosystems therefore requires treating verification semantics and release workflows as first-class usability targets rather than peripheral integration concerns.

Hieu Tran, Zhichao Yang, Zonghai Yao et al.

To enhance the performance of large language models (LLMs) in biomedical natural language processing (BioNLP) by introducing a domain-specific instruction dataset and examining its impact when combined with multi-task learning principles. We created the BioInstruct, comprising 25,005 instructions to instruction-tune LLMs(LLaMA 1 & 2, 7B & 13B version). The instructions were created by prompting the GPT-4 language model with three-seed samples randomly drawn from an 80 human curated instructions. We employed Low-Rank Adaptation(LoRA) for parameter-efficient fine-tuning. We then evaluated these instruction-tuned LLMs on several BioNLP tasks, which can be grouped into three major categories: question answering(QA), information extraction(IE), and text generation(GEN). We also examined whether categories(e.g., QA, IE, and generation) of instructions impact model performance. Comparing with LLMs without instruction-tuned, our instruction-tuned LLMs demonstrated marked performance gains: 17.3% in QA, 5.7% in IE, and 96% in Generation tasks. Our 7B-parameter instruction-tuned LLaMA 1 model was competitive or even surpassed other LLMs in the biomedical domain that were also fine-tuned from LLaMA 1 with vast domain-specific data or a variety of tasks. Our results also show that the performance gain is significantly higher when instruction fine-tuning is conducted with closely related tasks. Our findings align with the observations of multi-task learning, suggesting the synergies between two tasks. The BioInstruct dataset serves as a valuable resource and instruction tuned LLMs lead to the best performing BioNLP applications.

Chinh Ngo, Trieu H. Trinh, Long Phan et al.

We introduce MTet, the largest publicly available parallel corpus for English-Vietnamese translation. MTet consists of 4.2M high-quality training sentence pairs and a multi-domain test set refined by the Vietnamese research community. Combining with previous works on English-Vietnamese translation, we grow the existing parallel dataset to 6.2M sentence pairs. We also release the first pretrained model EnViT5 for English and Vietnamese languages. Combining both resources, our model significantly outperforms previous state-of-the-art results by up to 2 points in translation BLEU score, while being 1.6 times smaller.

Long Phan, Hieu Tran, Hieu Nguyen et al.

We present ViT5, a pretrained Transformer-based encoder-decoder model for the Vietnamese language. With T5-style self-supervised pretraining, ViT5 is trained on a large corpus of high-quality and diverse Vietnamese texts. We benchmark ViT5 on two downstream text generation tasks, Abstractive Text Summarization and Named Entity Recognition. Although Abstractive Text Summarization has been widely studied for the English language thanks to its rich and large source of data, there has been minimal research into the same task in Vietnamese, a much lower resource language. In this work, we perform exhaustive experiments on both Vietnamese Abstractive Summarization and Named Entity Recognition, validating the performance of ViT5 against many other pretrained Transformer-based encoder-decoder models. Our experiments show that ViT5 significantly outperforms existing models and achieves state-of-the-art results on Vietnamese Text Summarization. On the task of Named Entity Recognition, ViT5 is competitive against previous best results from pretrained encoder-based Transformer models. Further analysis shows the importance of context length during the self-supervised pretraining on downstream performance across different settings.

Hieu Tran, Son Nguyen, I-Ling Yen et al.

Vehicle arrival time prediction has been studied widely. With the emergence of IoT devices and deep learning techniques, estimated time of arrival (ETA) has become a critical component in intelligent transportation systems. Though many tools exist for ETA, ETA for special vehicles, such as ambulances, fire engines, etc., is still challenging due to the limited amount of traffic data for special vehicles. Existing works use one model for all types of vehicles, which can lead to low accuracy. To tackle this, as the first in the field, we propose a deep transfer learning framework TLETA for the driving time prediction. TLETA constructs cellular spatial-temporal knowledge grids for extracting driving patterns, combined with the road network structure embedding to build a deep neural network for ETA. TLETA contains transferable layers to support knowledge transfer between different categories of vehicles. Importantly, our transfer models only train the last layers to map the transferred knowledge, that reduces the training time significantly. The experimental studies show that our model predicts travel time with high accuracy and outperforms many state-of-the-art approaches.

Anh T. V. Dau, Hieu Trung Dao, Anh Tuan Nguyen et al.

Mainframe operating systems, despite their inception in the 1940s, continue to support critical sectors like finance and government. However, these systems are often viewed as outdated, requiring extensive maintenance and modernization. Addressing this challenge necessitates innovative tools that can understand and interact with legacy codebases. To this end, we introduce XMainframe, a state-of-the-art large language model (LLM) specifically designed with knowledge of mainframe legacy systems and COBOL codebases. Our solution involves the creation of an extensive data collection pipeline to produce high-quality training datasets, enhancing XMainframe's performance in this specialized domain. Additionally, we present MainframeBench, a comprehensive benchmark for assessing mainframe knowledge, including multiple-choice questions, question answering, and COBOL code summarization. Our empirical evaluations demonstrate that XMainframe consistently outperforms existing state-of-the-art LLMs across these tasks. Specifically, XMainframe achieves 30% higher accuracy than DeepSeek-Coder on multiple-choice questions, doubles the BLEU score of Mixtral-Instruct 8x7B on question answering, and scores six times higher than GPT-3.5 on COBOL summarization. Our work highlights the potential of XMainframe to drive significant advancements in managing and modernizing legacy systems, thereby enhancing productivity and saving time for software developers.

Zonghai Yao, Nandyala Siddharth Kantu, Guanghao Wei et al.

The advancement in healthcare has shifted focus toward patient-centric approaches, particularly in self-care and patient education, facilitated by access to Electronic Health Records (EHR). However, medical jargon in EHRs poses significant challenges in patient comprehension. To address this, we introduce a new task of automatically generating lay definitions, aiming to simplify complex medical terms into patient-friendly lay language. We first created the README dataset, an extensive collection of over 50,000 unique (medical term, lay definition) pairs and 300,000 mentions, each offering context-aware lay definitions manually annotated by domain experts. We have also engineered a data-centric Human-AI pipeline that synergizes data filtering, augmentation, and selection to improve data quality. We then used README as the training data for models and leveraged a Retrieval-Augmented Generation method to reduce hallucinations and improve the quality of model outputs. Our extensive automatic and human evaluations demonstrate that open-source mobile-friendly models, when fine-tuned with high-quality data, are capable of matching or even surpassing the performance of state-of-the-art closed-source large language models like ChatGPT. This research represents a significant stride in closing the knowledge gap in patient education and advancing patient-centric healthcare solutions.

Hieu Tran, Zonghai Yao, Junda Wang et al.

This work introduces RARE (Retrieval-Augmented Reasoning Enhancement), a versatile extension to the mutual reasoning framework (rStar), aimed at enhancing reasoning accuracy and factual integrity across large language models (LLMs) for complex, knowledge-intensive tasks such as commonsense and medical reasoning. RARE incorporates two innovative actions within the Monte Carlo Tree Search (MCTS) framework: A6, which generates search queries based on the initial problem statement, performs information retrieval using those queries, and augments reasoning with the retrieved data to formulate the final answer; and A7, which leverages information retrieval specifically for generated sub-questions and re-answers these sub-questions with the relevant contextual information. Additionally, a Retrieval-Augmented Factuality Scorer is proposed to replace the original discriminator, prioritizing reasoning paths that meet high standards of factuality. Experimental results with LLaMA 3.1 show that RARE enables open-source LLMs to achieve competitive performance with top open-source models like GPT-4 and GPT-4o. This research establishes RARE as a scalable solution for improving LLMs in domains where logical coherence and factual integrity are critical.

Hieu Tran, Ngoc Tran, Son Nguyen et al.

In modern Web technology, JavaScript (JS) code plays an important role. To avoid the exposure of original source code, the variable names in JS code deployed in the wild are often replaced by short, meaningless names, thus making the code extremely difficult to manually understand and analysis. This paper presents JSNeat, an information retrieval (IR)-based approach to recover the variable names in minified JS code. JSNeat follows a data-driven approach to recover names by searching for them in a large corpus of open-source JS code. We use three types of contexts to match a variable in given minified code against the corpus including the context of properties and roles of the variable, the context of that variable and relations with other variables under recovery, and the context of the task of the function to which the variable contributes. We performed several empirical experiments to evaluate JSNeat on the dataset of more than 322K JS files with 1M functions, and 3.5M variables with 176K unique variable names. We found that JSNeat achieves a high accuracy of 69.1%, which is the relative improvements of 66.1% and 43% over two state-of-the-art approaches JSNice and JSNaughty, respectively. The time to recover for a file or for a variable with JSNeat is twice as fast as with JSNice and 4x as fast as with JNaughty, respectively.

Hieu Tran, Zonghai Yao, Won Seok Jang et al.

Generative AI has demonstrated strong potential in healthcare, from clinical decision support to patient-facing chatbots that improve outcomes. A critical challenge for deployment is effective human-AI communication, where content must be both personalized and understandable. We introduce MedReadCtrl, a readability-controlled instruction tuning framework that enables LLMs to adjust output complexity without compromising meaning. Evaluations of nine datasets and three tasks across medical and general domains show that MedReadCtrl achieves significantly lower readability instruction-following errors than GPT-4 (e.g., 1.39 vs. 1.59 on ReadMe, p<0.001) and delivers substantial gains on unseen clinical tasks (e.g., +14.7 ROUGE-L, +6.18 SARI on MTSamples). Experts consistently preferred MedReadCtrl (71.7% vs. 23.3%), especially at low literacy levels. These gains reflect MedReadCtrl's ability to restructure clinical content into accessible, readability-aligned language while preserving medical intent, offering a scalable solution to support patient education and expand equitable access to AI-enabled care.

Hieu Tran, Junda Wang, Yujan Ting et al.

Large language models (LLMs) have shown remarkable capabilities in various natural language processing tasks, yet they often struggle with maintaining factual accuracy, particularly in knowledge-intensive domains like healthcare. This study introduces LEAF: Learning and Evaluation Augmented by Fact-Checking, a novel approach designed to enhance the factual reliability of LLMs, with a focus on medical question answering (QA). LEAF utilizes a dual strategy to enhance the factual accuracy of responses from models such as Llama 3 70B Instruct and Llama 3 8B Instruct. The first strategy, Fact-Check-Then-RAG, improves Retrieval-Augmented Generation (RAG) by incorporating fact-checking results to guide the retrieval process without updating model parameters. The second strategy, Learning from Fact-Checks via Self-Training, involves supervised fine-tuning (SFT) on fact-checked responses or applying Simple Preference Optimization (SimPO) with fact-checking as a ranking mechanism, both updating LLM parameters from supervision. These findings suggest that integrating fact-checked responses whether through RAG enhancement or self-training enhances the reliability and factual correctness of LLM outputs, offering a promising solution for applications where information accuracy is crucial.

Junda Wang, Yujan Ting, Eric Z. Chen et al.

Multimodal large language models (MLLMs) have made significant strides, yet they face challenges in the medical domain due to limited specialized knowledge. While recent medical MLLMs demonstrate strong performance in lab settings, they often struggle in real-world applications, highlighting a substantial gap between research and practice. In this paper, we seek to address this gap at various stages of the end-to-end learning pipeline, including data collection, model fine-tuning, and evaluation. At the data collection stage, we introduce SemiHVision, a dataset that combines human annotations with automated augmentation techniques to improve both medical knowledge representation and diagnostic reasoning. For model fine-tuning, we trained PMC-Cambrian-8B-AN over 2400 H100 GPU hours, resulting in performance that surpasses public medical models like HuatuoGPT-Vision-34B (79.0% vs. 66.7%) and private general models like Claude3-Opus (55.7%) on traditional benchmarks such as SLAKE and VQA-RAD. In the evaluation phase, we observed that traditional benchmarks cannot accurately reflect realistic clinical task capabilities. To overcome this limitation and provide more targeted guidance for model evaluation, we introduce the JAMA Clinical Challenge, a novel benchmark specifically designed to evaluate diagnostic reasoning. On this benchmark, PMC-Cambrian-AN achieves state-of-the-art performance with a GPT-4 score of 1.29, significantly outperforming HuatuoGPT-Vision-34B (1.13) and Claude3-Opus (1.17), demonstrating its superior diagnostic reasoning abilities.

Hieu Tran, Zonghai Yao, Lingxi Li et al.

Content generation conditioning on users's readability is an important application for personalization. In an era of large language models (LLMs), readability-controlled text generation based on LLMs has become increasingly important. This paper introduces a novel methodology called "Readability-Controlled Instruction Learning (ReadCtrl)," which aims to instruction-tune LLMs to tailor users' readability levels. Unlike the traditional methods, which primarily focused on categorical readability adjustments typically classified as high, medium, and low or expert and layperson levels with limited success, ReadCtrl introduces a dynamic framework that enables LLMs to generate content at various (near continuous level) complexity levels, thereby enhancing their versatility across different applications. Our results show that the ReadCtrl-Mistral-7B models significantly outperformed strong baseline models such as GPT-4 and Claude-3, with a win rate of 52.1%:35.7% against GPT-4 in human evaluations. Furthermore, Read-Ctrl has shown significant improvements in automatic evaluations, as evidenced by better readability metrics (e.g., FOG, FKGL) and generation quality metrics (e.g., BLEU, SARI, SummaC-Factuality, UniEval-Consistency and Coherence). These results underscore Read-Ctrl's effectiveness and tenacity in producing high-quality, contextually appropriate outputs that closely align with targeted readability levels, marking a significant advancement in personalized content generation using LLMs.

Nathan Beck, Abhiramon Rajasekharan, Hieu Tran

Transfer learning approaches in reinforcement learning aim to assist agents in learning their target domains by leveraging the knowledge learned from other agents that have been trained on similar source domains. For example, recent research focus within this space has been placed on knowledge transfer between tasks that have different transition dynamics and reward functions; however, little focus has been placed on knowledge transfer between tasks that have different action spaces. In this paper, we approach the task of transfer learning between domains that differ in action spaces. We present a reward shaping method based on source embedding similarity that is applicable to domains with both discrete and continuous action spaces. The efficacy of our approach is evaluated on transfer to restricted action spaces in the Acrobot-v1 and Pendulum-v0 domains. A comparison with two baselines shows that our method does not outperform these baselines in these continuous action spaces but does show an improvement in these discrete action spaces. We conclude our analysis with future directions for this work.

Hieu Trung Tran, Hanh Hong Phuc Vo, Son T. Luu

Finding a suitable job and hunting for eligible candidates are important to job seeking and human resource agencies. With the vast information about job descriptions, employees and employers need assistance to automatically detect job titles based on job description texts. In this paper, we propose the multi-label classification approach for predicting relevant job titles from job description texts, and implement the Bi-GRU-LSTM-CNN with different pre-trained language models to apply for the job titles prediction problem. The BERT with multilingual pre-trained model obtains the highest result by F1-scores on both development and test sets, which are 62.20% on the development set, and 47.44% on the test set.

Hieu Nguyen, Long Phan, James Anibal et al.

Text summarization is a challenging task within natural language processing that involves text generation from lengthy input sequences. While this task has been widely studied in English, there is very limited research on summarization for Vietnamese text. In this paper, we investigate the robustness of transformer-based encoder-decoder architectures for Vietnamese abstractive summarization. Leveraging transfer learning and self-supervised learning, we validate the performance of the methods on two Vietnamese datasets.

Hieu Tran, Son Nguyen, I-Ling Yen et al.

In this paper, we consider the IoT data discovery problem in very large and growing scale networks. Specifically, we investigate in depth the routing table summarization techniques to support effective and space-efficient IoT data discovery routing. Novel summarization algorithms, including alphabetical based, hash based, and meaning based summarization and their corresponding coding schemes are proposed. The issue of potentially misleading routing due to summarization is also investigated. Subsequently, we analyze the strategy of when to summarize in order to balance the tradeoff between the routing table compression rate and the chance of causing misleading routing. For experimental study, we have collected 100K IoT data streams from various IoT databases as the input dataset. Experimental results show that our summarization solution can reduce the routing table size by 20 to 30 folds with 2-5% increase in latency when compared with similar peer-to-peer discovery routing algorithms without summarization. Also, our approach outperforms DHT based approaches by 2 to 6 folds in terms of latency and traffic.

Hieu Tran, Long Phan, James Anibal et al.

In this paper, we propose SPBERT, a transformer-based language model pre-trained on massive SPARQL query logs. By incorporating masked language modeling objectives and the word structural objective, SPBERT can learn general-purpose representations in both natural language and SPARQL query language. We investigate how SPBERT and encoder-decoder architecture can be adapted for Knowledge-based QA corpora. We conduct exhaustive experiments on two additional tasks, including SPARQL Query Construction and Answer Verbalization Generation. The experimental results show that SPBERT can obtain promising results, achieving state-of-the-art BLEU scores on several of these tasks.

Hanh Hong-Phuc Vo, Hieu Trung Tran, Son T. Luu

Online game forums are popular to most of game players. They use it to communicate and discuss the strategy of the game, or even to make friends. However, game forums also contain abusive and harassment speech, disturbing and threatening players. Therefore, it is necessary to automatically detect and remove cyberbullying comments to keep the game forum clean and friendly. We use the Cyberbullying dataset collected from World of Warcraft (WoW) and League of Legends (LoL) forums and train classification models to automatically detect whether a comment of a player is abusive or not. The result obtains 82.69% of macro F1-score for LoL forum and 83.86% of macro F1-score for WoW forum by the Toxic-BERT model on the Cyberbullying dataset.

Long N. Phan, James T. Anibal, Hieu Tran et al.

In this report, we introduce SciFive, a domain-specific T5 model that has been pre-trained on large biomedical corpora. Our model outperforms the current SOTA methods (i.e. BERT, BioBERT, Base T5) on tasks in named entity relation, relation extraction, natural language inference, and question-answering. We show that text-generation methods have significant potential in a broad array of biomedical NLP tasks, particularly those requiring longer, more complex outputs. Our results support the exploration of more difficult text generation tasks and the development of new methods in this area

Hieu Tran, Cuong V. Dinh, Long Phan et al.

In this paper, we propose a Hierarchical Transformer model for Vietnamese spelling correction problem. The model consists of multiple Transformer encoders and utilizes both character-level and word-level to detect errors and make corrections. In addition, to facilitate future work in Vietnamese spelling correction tasks, we propose a realistic dataset collected from real-life texts for the problem. We compare our method with other methods and publicly available systems. The proposed method outperforms all of the contemporary methods in terms of recall, precision, and f1-score. A demo version is publicly available.

Long Phan, Hieu Tran, Daniel Le et al.

We present CoTexT, a pre-trained, transformer-based encoder-decoder model that learns the representative context between natural language (NL) and programming language (PL). Using self-supervision, CoTexT is pre-trained on large programming language corpora to learn a general understanding of language and code. CoTexT supports downstream NL-PL tasks such as code summarizing/documentation, code generation, defect detection, and code debugging. We train CoTexT on different combinations of available PL corpus including both "bimodal" and "unimodal" data. Here, bimodal data is the combination of text and corresponding code snippets, whereas unimodal data is merely code snippets. We first evaluate CoTexT with multi-task learning: we perform Code Summarization on 6 different programming languages and Code Refinement on both small and medium size featured in the CodeXGLUE dataset. We further conduct extensive experiments to investigate CoTexT on other tasks within the CodeXGlue dataset, including Code Generation and Defect Detection. We consistently achieve SOTA results in these tasks, demonstrating the versatility of our models.

Son Nguyen, Hoan Nguyen, Ngoc Tran et al.

Unexpected interactions among features induce most bugs in a configurable software system. Exhaustively analyzing all the exponential number of possible configurations is prohibitively costly. Thus, various sampling techniques have been proposed to systematically narrow down the exponential number of legal configurations to be analyzed. Since analyzing all selected configurations can require a huge amount of effort, fault-based configuration prioritization, that helps detect faults earlier, can yield practical benefits in quality assurance. In this paper, we propose CoPro, a novel formulation of feature-interaction bugs via common program entities enabled/disabled by the features. Leveraging from that, we develop an efficient feature-interaction aware configuration prioritization technique for a configurable system by ranking the configurations according to their total number of potential bugs. We conducted several experiments to evaluate CoPro on the ability to detect configuration-related bugs in a public benchmark. We found that CoPro outperforms the state-of-the-art configuration prioritization techniques when we add them on advanced sampling algorithms. In 78% of the cases, CoPro ranks the buggy configurations at the top 3 positions in the resulting list. Interestingly, CoPro is able to detect 17 not-yet-discovered feature-interaction bugs.