Realignment is a promising strategy to improve cross-lingual transfer in multilingual language models. However, empirical results are mixed and often unreliable, particularly for typologically distant or low-resource languages (LRLs) compared to English. Moreover, word realignment tools often rely on high-quality parallel data, which can be scarce or noisy for many LRLs. In this work, we conduct an extensive empirical study to investigate whether realignment truly benefits from using all available languages, or if strategically selected subsets can offer comparable or even improved cross-lingual transfer, and study the impact on LRLs. Our controlled experiments show that realignment can be particularly effective for LRLs and that using carefully selected, linguistically diverse subsets can match full multilingual alignment, and even outperform it for unseen LRLs. This indicates that effective realignment does not require exhaustive language coverage and can reduce data collection overhead, while remaining both efficient and robust when guided by informed language selection.

Ever wish you could listen to documents? In this post, we'll use machine learning to transform PDFs into audiobooks. This project was a collaboration with Kaz Sato. Update: Many of you have asked me what the total cost of this project is, which I've included at the end of this post. These days, you can do anything on foot: listen to the news, take meetings, even write notes (with voice dictation).

This project was originally designed by Kaz Sato. These days, you can do anything on foot: listen to the news, take meetings, even write notes (with voice dictation). The only thing you can't do while walking is read machine learning research papers. In this post, I'll show you how to use machine learning to transform documents in PDF or image format into audiobooks, using computer vision and text-to-speech. That way, you can read research papers on the go.

Technology is global, but ideas are local. The same IoT technology is being deployed all over the world, but a small Japanese startup might be who helps us make sense of it all. There is amazing work being done in user experience design, but most designers are operating with the contract of keeping users engaged. This is a fundamental shift from the traditional user-centered and functional design approaches. Today we sit down with Kaz Oki, founder of Mui Lab, and we talk about user design can actually improve our lives and help us disengage. We also talk about the challenges of getting VCs to invest in hardware startups, why Kyoto might be Japan's next innovation hub, and what it takes for a startup to successfully spin out of a Japanese company It's a great discussion, and I think you will really enjoy it. Welcome to Disrupting Japan, straight talk from Japan's most successful entrepreneurs. If you're a fan of Disrupting Japan, you know that I have a strong dislike for attempts to make Japan sound too exotic and this goes in both directions. On one side, we have consultants who claim that Japanese business practices are so unique, arcane, and confusing that the only way westerners can possibly understand them is by paying large sums of money to consultants such as themselves. And on the other side, of course, we have people insisting that foreigners can't really understand Japanese anime without a thorough and nuanced knowledge of Japanese language and history. I mean, there are differences, of course, and those differences should be acknowledged and respected, but whether an idea is coming from Japan or America, or Germany, one true measure of the value of that idea is its universality. The most important achievements might emerge out of cultural biases or sensitivities but they address something universally true, something deeply human. Today, we sit down with Kaz Oki of Mui Lab and we're going to talk about Mui's radical rethinking of how we should interact with computers and the different contexts for that interaction. The Mui itself is a tactile and visual user interface that literally fades into the furniture when you're not using it.