Technical and financial problems have been identified in the year since Japan's first transportation service using so-called Level 4 autonomous driving began in the town of Eiheiji, Fukui Prefecture. Amid the country's declining population, Level 4 autonomous driving, or driving that is fully automated under certain conditions, is viewed as a promising means of transport. The service in Eiheiji, however, has shown the hurdles that must be cleared. On May 28, 2023, the service was launched on a 2-kilometer section of a walkway in Eiheiji. It is available only on Saturdays, Sundays and national holidays.

The vast majority of the popular English named entity recognition (NER) datasets contain American or British English data, despite the existence of many global varieties of English. As such, it is unclear whether they generalize for analyzing use of English globally. To test this, we build a newswire dataset, the Worldwide English NER Dataset, to analyze NER model performance on low-resource English variants from around the world. We test widely used NER toolkits and transformer models, including models using the pre-trained contextual models RoBERTa and ELECTRA, on three datasets: a commonly used British English newswire dataset, CoNLL 2003, a more American focused dataset OntoNotes, and our global dataset. All models trained on the CoNLL or OntoNotes datasets experienced significant performance drops-over 10 F1 in some cases-when tested on the Worldwide English dataset. Upon examination of region-specific errors, we observe the greatest performance drops for Oceania and Africa, while Asia and the Middle East had comparatively strong performance. Lastly, we find that a combined model trained on the Worldwide dataset and either CoNLL or OntoNotes lost only 1-2 F1 on both test sets.

The nation's first vehicle accident involving so-called Level 4 autonomous driving occurred in October due to a failure to visually recognize an object, a local government report said. The report was released recently by the town of Eiheiji, Fukui Prefecture, where Level 4 driving, or fully automated driving under certain conditions, received approval, the first local government to do so in the country. The town government plans to improve Level 4 vehicles' visual recognition performance to prevent any similar accidents. In the accident in the town on Oct. 29, the autonomous vehicle, traveling at about 4 kilometers per hour, hit a pedal of a parked bicycle. The vehicle detected the crash and made an emergency stop.

Japan is pushing for 50 locations with driverless services in place within three years, but fully autonomous vehicles remain nearly nonexistent in the country. So far, Fukui Prefecture is the only place with vehicles featuring level-4 capabilities -- defined when they can handle all driving tasks -- but only under specific conditions with the option for humans to take over. In the town of Eiheiji, the seven-seater golf carts are only allowed to navigate a 2 kilometer course. The limited availability of autonomous driving in Japan stands in stark contrast to the U.S. and China, where robotaxis already roam the streets in some cities. Waymo, backed by Google parent Alphabet, and General Motor's Cruise are testing driverless taxi services in San Francisco.

Such services are expected to become a new means of public transit in regions facing population decline. In Eiheiji, where level 4 autonomous driving was approved for the first time in the country, a seven-seater electric cart developed by the National Institute of Advanced Industrial Science and Technology and others runs on a section of a walking trail spanning about 2 kilometers. There is no operator in the cart, and one person in charge of remote monitoring manages up to three such electric carts. This could be due to a conflict with your ad-blocking or security software. Please add japantimes.co.jp and piano.io to your list of allowed sites.

Fukui – The town of Eiheiji in Fukui Prefecture, known for its ancient Zen Buddhist temple, is set to become the first place in the country to adopt a new type of autonomous self-driving vehicle, hoping to pioneer investment in what could become an important future technology. According to the transport ministry, "level-4" self-driving autonomous vehicles are slated to begin operating in the town from fiscal 2023, which starts in April, marking the first time for authorities in Japan to give the go-ahead for such a project. Autonomous vehicles are expected to become an essential means of transportation in regions of the country where public transport is becoming increasingly scarce. The town, which considers itself to be at the forefront of offering solutions, will aim to identify and solve operational issues for the widespread adoption of driverless technology. This could be due to a conflict with your ad-blocking or security software.

Demonstration-based learning has shown great potential in stimulating pretrained language models' ability under limited data scenario. Simply augmenting the input with some demonstrations can significantly improve performance on few-shot NER. However, why such demonstrations are beneficial for the learning process remains unclear since there is no explicit alignment between the demonstrations and the predictions. In this paper, we design pathological demonstrations by gradually removing intuitively useful information from the standard ones to take a deep dive of the robustness of demonstration-based sequence labeling and show that (1) demonstrations composed of random tokens still make the model a better few-shot learner; (2) the length of random demonstrations and the relevance of random tokens are the main factors affecting the performance; (3) demonstrations increase the confidence of model predictions on captured superficial patterns. We have publicly released our code at https://github.com/SALT-NLP/RobustDemo.

A suicide prevention group will dispatch a drone to monitor remote areas around Tojinbo in Fukui Prefecture -- whose lonely cliffs remain romanticized in popular imagination as a destination where people go to end their lives -- in the hopes that the technology will enhance efforts to minimize the suicide rate. Retired police officer Yukio Shige, the 73-year-old head of the nonprofit group, said that although there has been a decline in the total number of people leaping off the cliffs in recent years, suicide remains a persistent problem. "This year we have managed to stop five people from committing suicide but five is a very small number; it's only one per month," Shige said, adding that far more still think about killing themselves. "So far we've monitored (Tojinbo) by ourselves but with the use of drones we could reach places that the human eye can't see." The group's 16 members -- made up of retired police officers, academics and company workers -- patrol the cliffs six times a week, from 11 a.m. until sunset.

At a Tokyo job fair for the atomic energy industry on March 4, Kenta Kakitani, a graduate student at the University of Tokyo, hopes to some day become a nuclear plant design engineer. But Kakitani may be a rare breed in Japan, where nuclear businesses have seen a serious shortage of new talent since the March 11, 2011, meltdowns at the Fukushima No. 1 power plant, the world's worst nuclear disaster since Chernobyl in 1986. "It seems that the nuclear power industry has lost much of its popularity because it is seen as in decline and is suffering a negative image from having to decommission crippled reactors," said Kakitani, 24, who majors in nuclear engineering. According to education ministry data, 298 students entered departments related to nuclear power study in fiscal 2015, a slight decline from 317 in fiscal 2010. Kakitani said that although the number may not have declined drastically, many talented students are majoring in the fields of artificial intelligence and aerospace engineering instead of nuclear engineering. The turnout at the job fair reflects the nuclear power industry's fall from grace.

With improvements in facial-recognition technology and the increasing popularity of smartphones, the threat to one's privacy unexpectedly posed by Internet photos posted by strangers is growing day by day. To protect against unwanted scrutiny, trading company Nissey Corp., based in Sabae, Fukui Prefecture, has developed special glasses that thwart electronic facial recognition. The Privacy Visor is a set of white titanium goggles with lenses that have a mesh-like surface providing many tiny spaces to see through. "Since the view might not be clear, driving or riding on bicycles must be avoided," a Nissey spokesman warned. The glasses were jointly developed with the National Institute of Informatics, a research institute dedicated to information technology.