An open-source traditional Mandarin large language model (LLM) localized for Taiwan was released yesterday, achieving higher scores in the Taiwanese bar exam than GPT-4o.
The project, called Taiwan Mixture of Experts (TAME), was initiated by Chang Chun Group, Pegatron, Unimicron, Chang Gung Memorial Hospital, TechOrange and National Taiwan University.
Using Nvidia’s Taipei-1 supercomputer in Kaohsiung, it was trained on nearly 500 billion tokens of data provided by the partners in their specialized fields.
Photo: Reuters
On its first attempt taking last year’s bar exam, Project TAME did better than 89 percent of test takers, LegalSign.ai chief executive officer Steven Chen (陳啟桐) said.
After only a few months, it has already received outstanding marks when taking exams for university students, legal professionals, traditional Chinese medicine practitioners, tour guides and drivers, Project TAME said.
Out of 39 evaluations and nearly 3,000 questions, Project TAME outperformed all other available models, with an accuracy rate 6.8 percent higher than the second-place Claude-Opus and 9.3 percent higher than GPT-4o.
The model is available to use for free online at https://twllm.com and would be open source, organizers said.
An interactive chatbot version for the public was created by AMPIC.
It is also available for download by enterprises and developers on Github.
The new LLM is the result of 350,000 GPU hours and 1,285 working hours by 31 engineers, Project TAME leader and Pegatron associate vice president of AI development Andrew Hsiao (蕭安助) said.
Its two specialties are localization for Taiwan and application for industry, he said.
It was only released today, but enterprises would be able to verify and fine-tune the model as time goes on to verify whether it is suitable for use in different fields, he added.
WANG RELEASED: A police investigation showed that an organized crime group allegedly taught their clients how to pretend to be sick during medical exams Actor Darren Wang (王大陸) and 11 others were released on bail yesterday, after being questioned for allegedly dodging compulsory military service or forging documents to help others avoid serving. Wang, 33, was catapulted into stardom for his role in the coming-of-age film Our Times (我的少女時代). Lately, he has been focusing on developing his entertainment career in China. The New Taipei District Prosecutors’ Office last month began investigating an organized crime group that is allegedly helping men dodge compulsory military service using falsified documents. Police in New Taipei City Yonghe Precinct at the end of last month arrested the main suspect,
A cat named Mikan (蜜柑) has brought in revenue of more than NT$10 million (US$305,390) for the Kaohsiung MRT last year. Mikan, born on April 4, 2020, was a stray cat before being adopted by personnel of Kaohsiung MRT’s Ciaotou Sugar Refinery Station. Mikan was named after a Japanese term for mandarin orange due to his color and because he looks like an orange when curled up. He was named “station master” of Ciaotou Sugar Refinery Station in September 2020, and has since become famous. With Kaohsiung MRT’s branding, along with the release of a set of cultural and creative products, station master Mikan
RISING TOURISM: A survey showed that tourist visits increased by 35 percent last year, while newly created attractions contributed almost half of the growth Changhua County’s Lukang Old Street (鹿港老街) and its surrounding historical area clinched first place among Taiwan’s most successful tourist attractions last year, while no location in eastern Taiwan achieved a spot in the top 20 list, the Tourism Administration said. The listing was created by the Tourism Administration’s Forward-looking Tourism Policy Research office. Last year, the Lukang Old Street and its surrounding area had 17.3 million visitors, more than the 16 million visitors for the Wenhua Road Night Market (文化路夜市) in Chiayi City and 14.5 million visitors at Tainan’s Anping (安平) historical area, it said. The Taipei 101 skyscraper and its environs —
Taiwan on Friday said a New Zealand hamburger restaurant has apologized for a racist remark to a Taiwanese customer after reports that it had first apologized to China sparked outrage in Taiwan. An image posted on Threads by a Taiwanese who ate at Fergburger in Queenstown showed that their receipt dated Sunday last week included the words “Ching Chang,” a racial slur. The Chinese Consulate-General in Christchurch in a statement on Thursday said it had received and accepted an apology from the restaurant over the incident. The comment triggered an online furor among Taiwanese who saw it as an insult to the
RSS