Vietnamese
PulseAugur coverage of Vietnamese — every cluster mentioning Vietnamese across labs, papers, and developer communities, ranked by signal.
6 day(s) with sentiment data
-
LLMs struggle with Vietnamese dialects, new benchmark reveals
A new benchmark, VialectBench, has been developed to evaluate the robustness of Large Language Models (LLMs) to Vietnamese dialects. The benchmark includes 2,400 dialectal rewrites across six dialect groups for tasks su…
-
yingsuan.top tool generates Thai/Vietnamese replies in seconds
A new multilingual customer service tool from yingsuan.top has been tested for its ability to generate rapid responses in Thai and Vietnamese for cross-border e-commerce stores. The tool aims to solve the problem of slo…
-
New benchmark standardizes machine learning for mental health classification
Researchers have introduced the Neurai-VN Benchmark, a standardized framework for evaluating machine learning models in mental health classification using digital phenotyping. This benchmark is built upon the Neurai-VN …
-
New MERaLiON-GR model achieves state-of-the-art gender recognition across multiple languages
Researchers have developed MERaLiON-GR, a novel speech gender recognition model capable of classifying gender for both English and several Southeast Asian languages. This model is built upon MERaLiON-SpeechEncoder-2, a …
-
LLM safety weaker in lower-resource languages, audit finds
A recent audit of the Qwen3-30B-A3B model revealed that its safety alignment is weaker in lower-resource languages compared to English and Standard Chinese. Using an automated auditing framework called Petri, researcher…
-
New method bridges Bahnaric-Vietnamese translation gap
Researchers have developed a method for Bahnaric-Vietnamese translation, aiming to bridge cultural understanding between ethnic groups in Vietnam. Due to a scarcity of Bahnaric language resources, the team employed a tr…
-
New benchmarks and methods advance medical vision-language models
Researchers have developed new benchmarks and distillation techniques to improve the capabilities of vision-language models (VLMs) in the medical domain. PathAgentBench focuses on evaluating VLMs' ability to acquire and…
-
SpiS-GAN generates realistic handwriting, improving recognition systems · 2 sources tracked
Researchers have developed SpiS-GAN, a novel framework for synthesizing realistic handwriting to address the scarcity of annotated data for training handwriting recognition systems. This generative adversarial network u…
-
New BamiBERT model sets Vietnamese language benchmark
Researchers have developed BamiBERT, a new language model specifically designed for Vietnamese. This model, trained on a large corpus, offers an extended context length of 2048 tokens and processes raw text without requ…
-
LLMs face persistent plasticity loss despite scaling, study finds
A new arXiv paper investigates the challenge of plasticity loss in large language models (LLMs), where models struggle to learn new information after being trained on older data. Researchers found that this plasticity l…
-
New VieSpeaker Dataset Enhances Vietnamese Speaker Recognition Without Visual Cues · 3 sources tracked
Researchers have introduced VieSpeaker, a new large-scale dataset for Vietnamese speaker recognition that does not rely on visual cues. This dataset was constructed using a novel pipeline that leverages textual metadata…
-
Malaysia's blue economy ambitions threatened by maritime security gaps
Malaysia's aspirations for a thriving blue economy are being hampered by significant security gaps at sea. Local fishermen in Terengganu have reported incursions by Vietnamese vessels into Malaysian waters, particularly…
-
New CzechDocs dataset aids format-preserving machine translation research
Researchers have introduced CzechDocs, a new dataset designed for evaluating machine translation systems that preserve document formatting. This multiway parallel dataset includes documents in Czech and several minority…
-
User configures EVi with SUSE vimrc and plans openSUSE packaging
The user has configured EVi locally, setting the SUSE vimrc as the default. They have also created aliases for Vim and vi to point to EVi, intending to manage muscle memory. The next step involves packaging EVi for open…
-
New fashion datasets boost AI sketch-to-image and retrieval capabilities
Two new large-scale datasets, GarmentSketch and VietFashion, have been released to advance sketch-to-image generation and retrieval in the fashion industry. GarmentSketch offers 26,249 fashion sketches paired with detai…
-
New technique boosts Vietnamese speech translation accuracy
Researchers have developed a new data augmentation technique called Phonetically-Informed Data Augmentation (PiDA) to improve Vietnamese speech translation. The method addresses error propagation in cascaded speech tran…
-
Meister Jeder's AI Portraits Explore Identity Facades · 10 sources tracked
Photographer Meister Jeder is creating AI-generated portraits that explore themes of identity and facade. These images, presented as "purely fake (AI) Photography," are accompanied by hashtags related to Dadaism and AI …
-
New Vietnamese ASR uses phoneme-based syllabic modeling
Researchers have developed a novel Syllabic-Structure Decoder for Automatic Speech Recognition (ASR) systems specifically for Vietnamese. This new approach models speech at the phoneme level, explicitly capturing the ph…
-
Vietnamese speech model captures dialectal phonetic variations
Researchers have developed a new framework to model phonetic variations in Vietnamese speech across different regions. This approach explicitly accounts for dialectal differences at both the vocabulary and decoding stag…
-
Indonesia becomes new hub for scam syndicates after police raids
Indonesian authorities have recently arrested hundreds of foreign nationals in a crackdown on online scam and gambling operations. These crime groups appear to be shifting their base of operations to Indonesia, attracte…