Recent advances in large language models (LLMs) have yielded impressive performance on various tasks, yet they often depend on high-quality feedback that can be costly. Self-refinement methods attempt to leverage LLMs' internal evaluation mechanisms with minimal human supervision; however, these approaches frequently suffer from inherent biases and overconfidence, especially in domains where the models lack sufficient internal knowledge, resulting in performance degradation. As an initial step toward enhancing self-refinement for broader applications, we introduce an iterative refinement pipeline that employs the Unlabeled-Unlabeled learning framework to improve LLM-generated pseudo-labels for classification tasks.

Online structured prediction is a task of sequentially predicting outputs with complex structures based on inputs and past observations, encompassing online classification. Recent studies showed that in the full-information setting, we can achieve finite bounds on the surrogate regret, i.e., the extra target loss relative to the best possible surrogate loss. In practice, however, full-information feedback is often unrealistic as it requires immediate access to the whole structure of complex outputs. Motivated by this, we propose algorithms that work with less demanding feedback, bandit and delayed feedback. For bandit feedback, by using a standard inverseweighted gradient estimator, we achieve a surrogate regret bound of O( KT) for the time horizon T and the size of the output set K. However, K can be extremely large when outputs are highly complex, resulting in an undesirable bound. To address this issue, we propose another algorithm that achieves a surrogate regret bound of O(T2/3), which is independent of K. This is achieved with a carefully designed pseudo-inverse matrix estimator. Furthermore, we numerically compare the performance of these algorithms, as well as existing ones. Regarding delayed feedback, we provide algorithms and regret analyses that cover various scenarios, including full-information and bandit feedback, as well as fixed and variable delays.

We present The Matrix, a foundational realistic world simulator capable of generating infinitely long 720p high-fidelity real-scene video streams with real-time, responsive control in both first-and third-person perspectives. Trained on limited supervised data from video games like Forza Horizon 5 and Cyberpunk 2077, complemented by large-scale unsupervised footage from real-world settings like Tokyo streets, The Matrix allows users to traverse diverse terrains--deserts, grasslands, water bodies, and urban landscapes--in continuous, uncut hour-long sequences. With speeds of up to 16 FPS, the system supports real-time interactivity and demonstrates zero-shot generalization, translating virtual game environments to real-world contexts where collecting continuous movement data is often infeasible. For example, The Matrix can simulate a BMW X3 driving through an office setting--an environment present in neither gaming data nor real-world sources. This approach showcases the potential of game data to advance robust world models, bridging the gap between simulations and real-world applications in scenarios with limited data.

A group of high school students arrested over allegedly trying to extort money from a boy in western Tokyo may have used ChatGPT to decide how much to demand, police said. A group of high school students in Tokyo arrested over allegedly assaulting a boy and trying to extort money from him may have used ChatGPT to decide how much to demand, media reports have recently revealed. Five teenagers, including a 17-year-old girl and four boys ranging in age from 16 to 17, were arrested in January over the alleged assault and attempted extortion of a 17-year-old high school student in the city of Hachioji in western Tokyo, according to the Metropolitan Police Department. Police said the suspects assaulted the boy in a plaza in Hachioji's Shiroyamate district, breaking his nose and causing other injuries, before allegedly trying to extort ¥150,000 ($935) from him. The girl, who was the victim's ex-girlfriend, allegedly first confronted him, accusing him of touching her younger sister's leg. She then challenged him, saying, "Give me the money or fight me one-on-one," according to reports by Fuji TV.

Dipu Tamang from Nepal is among more than 400,000 international students in Japan. When Dipu Tamang arrived in Japan from Nepal in 2024, he joined a growing stream of young people who see the country less as a traditional study destination and more as a structured route into work and long-term opportunity. The 22-year-old graduated from Shinjuku Heiwa Japanese Language School in March and now studies international business at a vocational college in Tokyo. He juggles part-time work as a convenience store clerk and hotel housekeeper to help cover his living expenses. "At first, I was interested in Japanese pop culture," he said. "Then I wanted to learn the language.

Manatomo Yoneyama, president of Sumitomo Mitsui Trust Bank, speaks during an interview at the bank's headquarters in Tokyo's Chiyoda Ward on May 22. | JIJI Sumitomo Mitsui Trust Bank plans to invest ¥360 billion ($2.2 billion) to ¥380 billion in digital technologies over three years, President Manatomo Yoneyama said in an interview. The bank also plans to spend ¥30 billion to optimize its operations. It will utilize artificial intelligence technology for office tasks and reposition some 900 employees to client-facing roles. The bank made an AI agent, which can handle people's tasks, "100% internally," said Yoneyama, who took the helm of Sumitomo Mitsui Trust in April after working on digital innovation at the bank. He said the bank is "sensing the advantage" of the in-house development, including smooth utilization of data. It aims to sell the AI agent in fiscal 2028.

Junya Ogawa, leader of the Centrist Reform Alliance, attends a news conference on Friday in Tokyo. Moves to launch a new party have emerged among three opposition parties in Japan, sources said Sunday. The idea arose in the course of talks on the possible integration of the Constitutional Democratic Party of Japan and Komeito into the Centrist Reform Alliance. The CRA and Komeito are keen about the new opposition party option. Some senior members of the CDP and officials of a major labor union supporting the initiative are also positive, according to sources familiar with the matter. In a time of both misinformation and too much information, quality journalism is more crucial than ever.

Sakie Yokota (center, back), mother of North Korean abductee Megumi Yokota, and others attend a rally held in Tokyo on Saturday that called for the immediate return of Japanese people abducted by North Korea. A large-scale rally was held in Tokyo on Saturday to seek the immediate return home of all Japanese abductees in North Korea. Relatives of those abducted to North Korea decades ago expressed hopes for the return of abductees immediately and while their parents are still alive. The event, organized by the association of families of abduction victims and other entities, was attended by about 800 people, including Prime Minister Sanae Takaichi. "We will never give up," said Takuya Yokota, 57, head of the association and the younger brother of Megumi Yokota, who was abducted in 1977 at the age of 13. He called on North Korean leader Kim Jong Un to release all abductees to "chart a bright future for both countries."

A humanoid robot is displayed at the Humanoids Summit in Tokyo on Thursday. Amid growing anticipation of physical artificial intelligence, robots are increasingly being seen as a viable option to fill labor gaps. This topic was one of the major agendas during the two-day Humanoids Summit in Tokyo, which ended on Friday. "In Japan the United States globally, we just don't have the birth rates to sustain the workforce that we need," said Brendan Schulman, vice president of policy at Massachusetts-based robotics company Boston Dynamics during a speech at the event. In a time of both misinformation and too much information, quality journalism is more crucial than ever.

Taiyo Yuden sees'scary' levels of AI parts demand risking supply chain Multilayer ceramic capacitors, which are tiny components that regulate and stabilize power flow in electronic devices, are becoming a growing bottleneck in the construction of artificial intelligence data centers. Taiyo Yuden is fielding "scary" levels of demand for its high-end artificial intelligence server components, stretching capacity and increasing the risk of supply chain hiccups. The Tokyo-based company, which makes multilayer ceramic capacitors, will likely need to accelerate spending to expand output capacity, Chief Executive Officer Katsuya Sase said in an interview. MLCCs, which are tiny components that regulate and stabilize power flow in electronic devices, are becoming a growing bottleneck in the construction of artificial intelligence data centers. Taiyo Yuden and Murata Manufacturing comprise the bulk of the world's supplies of high-end MLCCs. "The volumes we are seeing today -- it's scary," Sase said.