Which Works Best for Vietnamese? A Practical Study of Information Retrieval Methods across Domains (2026.findings-eacl)
Copied to clipboard
| Challenge: | Existing studies on Large Language Models (LLMs) are limited to single domains or curated datasets. |
| Approach: | They propose a domain-normalized, multi-domain benchmark for Vietnamese IR . they evaluate lexical, neural-sparse, late-interaction, dense, and hybrid paradigms . |
| Outcome: | The proposed benchmarks cover six domains and ten datasets across education, legal, healthcare, customer support, lifestyle reviews, and open-domain knowledge. |
Similar Papers
ViGPTQA - State-of-the-Art LLMs for Vietnamese Question Answering: System Overview, Core Models Training, and Evaluations (2023.emnlp-industry)
Copied to clipboard
| Challenge: | Large language models (LLMs) and their applications in low-resource languages are limited due to lack of training data and benchmarking datasets. |
| Approach: | They propose a question-response system for Vietnamese that uses LLMs . they propose to open-source the model and train it on benchmark datasets based on Vietnamese data . |
| Outcome: | The proposed question answering system for Vietnamese is open-source and performant . it can learn and capture human-like text, but there is a gap in evaluations for Vietnamese . |
Improving Vietnamese-English Cross-Lingual Retrieval for Legal and General Domains (2025.naacl-short)
Copied to clipboard
Toan Ngoc Nguyen, Nam Le Hai, Nguyen Doan Hieu, Dai An Nguyen, Linh Ngo Van, Thien Huu Nguyen, Sang Dinh
| Challenge: | Existing document retrieval systems focus on a single language, targeting resource-rich languages like English or Chinese. |
| Approach: | They propose auxiliary loss function and symmetrical training strategy for cross-lingual retrieval between Vietnamese and English . they propose a dataset that covers the general domain and extends to the legal field . |
| Outcome: | The proposed dataset significantly improves state-of-the-art models on cross-lingual retrieval tasks. |
VMLU Benchmarks: A comprehensive benchmark toolkit for Vietnamese LLMs (2025.acl-long)
Copied to clipboard
Cuc Thi Bui, Nguyen Truong Son, Truong Van Trang, Lam Viet Phung, Pham Nhut Huy, Hoang Anh Le, Quoc Huu Van, Phong Nguyen-Thuan Do, Van Le Tran Truc, Duc Thanh Chau, Le-Minh Nguyen
| Challenge: | The evolution of Large Language Models (LLMs) has underscored the need for benchmarks designed for various languages and cultural contexts. |
| Approach: | They propose to use Vietnamese multitask language understanding (VMLU) benchmarks to assess different capabilities of LLMs, including general knowledge, reading comprehension, reasoning, and conversational skills. |
| Outcome: | The VMLU Benchmarks assess LLMs' general knowledge, reading comprehension, reasoning, and conversational skills. |
VIMQA: A Vietnamese Dataset for Advanced Reasoning and Explainable Multi-hop Question Answering (2022.lrec-1)
Copied to clipboard
| Challenge: | Existing Vietnamese Question Answering (QA) datasets do not explore the model’s ability to perform advanced reasoning and provide evidence to explain the answer. |
| Approach: | They propose to use Vietnamese as a question-answer dataset with 10,000 Wikipedia-based multi-hop question-and-answ pairs to test model's ability to reason and explain the answer. |
| Outcome: | The proposed dataset is in Vietnamese, a low-resource language. |
A Large-Scale Benchmark for Vietnamese Sentence Paraphrases (2025.findings-naacl)
Copied to clipboard
| Challenge: | 1.2M original–paraphrase pairs were generated using a hybrid approach to generate high-quality paraphrases. |
| Approach: | They present a high-quality Vietnamese dataset for sentence paraphrasing . they used automatic paraphrase generation and manual evaluation to ensure high quality . |
| Outcome: | The proposed dataset is the first large-scale study on Vietnamese paraphrasing . it combines automatic paraphrase generation with manual evaluation to ensure high quality . |
MKQA: A Linguistically Diverse Benchmark for Multilingual Open Domain Question Answering (2021.tacl-1)
Copied to clipboard
| Challenge: | Existing multilingual QA datasets lack linguistic diversity and comparable evaluation between languages. |
| Approach: | They propose a multilingual question-answer evaluation set with 10k English queries and human translations of them into 25 additional languages and dialects. |
| Outcome: | The proposed model is based on a multilingual knowledge questions and answers evaluation set with 26 languages. |
Exploring Language Model Generalization in Low-Resource Extractive QA (2025.coling-main)
Copied to clipboard
| Challenge: | Existing LLMs struggle with dataset demands of closed domains such as medicine and law . current LLM performance in closed domain is lacking, even on traditional tasks such as Natural Language Inference . |
| Approach: | They investigate Extractive Question Answering (EQA) with Large Language Models (LLMs) under domain drift . they find that LLMs struggle with dataset demands of closed domains . |
| Outcome: | The proposed model performs poorly in extractive question answering tasks under domain drift . the proposed model can generalize to domains that require specific knowledge without training . |
Crossing Linguistic Horizons: Finetuning and Comprehensive Evaluation of Vietnamese Large Language Models (2024.findings-naacl)
Copied to clipboard
| Challenge: | Existing open-source LLMs exhibit limited effectiveness in processing Vietnamese . lack of systematic benchmark datasets and metrics tailored for Vietnamese LLM evaluation exacerbates these issues. |
| Approach: | They propose to fine tune LLMs specifically for Vietnamese and develop a framework for evaluation . they find that larger models introduce more biases and uncalibrated outputs . |
| Outcome: | The proposed framework finetunes LLMs specifically for Vietnamese and provides a framework for evaluation . |
Comprehensive Comparison of RAG Methods Across Multi-Domain Conversational QA (2026.eacl-srw)
Copied to clipboard
| Challenge: | Existing studies evaluate RAG methods in isolation and focus on single-turn settings. |
| Approach: | They compare retrieval-augmented generation methods for multi-turn conversational QA with those that use dialogue history and coreference to ground large language models. |
| Outcome: | The proposed methods outperform vanilla RAG and advanced methods fail to yield gains and can even degrade performance below the No-RAG baseline. |
ViMedAQA: A Vietnamese Medical Abstractive Question-Answering Dataset and Findings of Large Language Model (2024.acl-srw)
Copied to clipboard
| Challenge: | Existing abstractive question-answering datasets in Vietnamese are lacking . |
| Approach: | They propose to introduce a Vietnamese abstractive question-answering corpus to address this gap . they propose to use Vietnamese abstractives to generate answers to questions . |
| Outcome: | The proposed dataset examines the capability of large language models in the Vietnamese medical domain, including reasoning, memorizing and awareness of essential information. |