Alibaba International has launched an upgraded version of its AI-powered translation tool, claiming it surpasses similar products from Google, DeepL, and OpenAI’s ChatGPT. The company claims that Alibaba’s new AI translation tool beats Google and ChatGPT in supporting seamless communication for e-commerce merchants. The tool, called Marco MT, has been benchmarked against Flores, a widely recognized translation framework, according to the Chinese e-commerce giant.
This release builds upon the initial version of the tool, introduced last year, which already has 500,000 users globally. Merchants can use the tool to translate product pages into the language of their target market, making cross-border sales more efficient.
Powered by Large Language Models
Marco MT is driven by large language models (LLMs), which use massive amounts of data to understand and generate human-like text. Unlike traditional translation tools, LLMs can detect and apply contextual clues, such as cultural or industry-specific terms, to provide more accurate translations. Kaifu Zhang, vice president of Alibaba International Digital Commerce Group, said this new version aims to help merchants increase sales by enhancing communication with international customers.
Alibaba’s new AI translation tool beats Google and ChatGPT in both speed and accuracy across 15 languages. Alibaba’s translation tool is powered by the company’s in-house model, Qwen, and supports 15 languages, including English, Chinese, French, German, and Japanese. Zhang expects substantial demand for the tool, particularly from Europe, the Americas, and emerging markets.
Supporting Global E-commerce
Since its initial launch, Alibaba’s AI translation tool has helped merchants translate over 100 million product listings. The updated tool allows sellers to communicate better with buyers by improving the quality of product descriptions, titles, and search keywords. It also helps in managing customer service interactions and search guidance, boosting the overall experience for both merchants and buyers.
Alibaba’s international platforms like AliExpress and Lazada are among the first to benefit from the enhanced tool. The company envisions further applications for the tool beyond e-commerce, potentially offering it to businesses in other sectors.
Tackling Translation Challenges
Many existing translation tools face difficulties in handling complex expressions, idiomatic language, and cultural nuances. Alibaba’s Marco MT addresses these issues, especially for languages that are less commonly used. The tool has performed well in Flores benchmarks, proving its ability to provide high-quality translations even for low-resource languages.
Alibaba aims to make Marco MT available not just for e-commerce but for broader use by individuals, small businesses, and larger enterprises. It plans to integrate the tool into its long-term AI strategy, helping businesses overcome language barriers and engage with customers across different regions.
Expanding Global Reach
The latest version of Marco MT aligns with Alibaba’s goal of expanding its global footprint. The company intends to offer the tool worldwide, helping businesses create content, manage customer service, and improve communication in multiple languages. Businesses using cross-border platforms have found that Alibaba’s new AI translation tool beats Google and ChatGPT in translating product descriptions and search keywords.
As Alibaba continues to explore new technologies, this translation tool is set to become a cornerstone of the company’s AI strategy, empowering businesses to expand their reach in international markets.
Alibaba’s new AI translation tool beats Google and ChatGPT in several important areas, making it a promising solution for businesses engaged in cross-border e-commerce. One of the key strengths of the tool is its ability to handle contextual translation. This means it goes beyond just translating words by considering cultural differences and industry-specific language, which is crucial for businesses looking to appeal to global markets.
The use of large language models (LLMs) allows Alibaba to train its tool on massive amounts of data, improving accuracy and making it more adaptable to different regions and languages. This has been particularly helpful for low-resource languages, where translation tools often struggle. By outperforming its competitors, Alibaba’s tool offers a competitive advantage to merchants who rely on accurate translations for product listings and customer interactions.
Also Read: TSMC is Set to Report Strong Profits Driven by AI Boom, Analysts Predict.
