tkgally

29 ms·

301.
by tkgally 2y ago
I tried the three-way conversation feature. While not perfect, it worked pretty well. I posted a demo here: https://youtu.be/pU_-Xf60DAU
302.
by tkgally 2y ago
I tried accessing Gemini 2.0 Flash through Google AI Studio in the Safari browser on my iPhone, and to my surprise it worked. After I gave it access to my microphone and camera, I was able to have a pretty smooth conversation with it about
303.
by tkgally 2y ago
As is her Instagram: https://www.instagram.com/depthsofwikipedia
304.
by tkgally 2y ago
A defining sentence at the top of the page, something like “Tree Calculus is a [noun phrase] for [summary of purpose],” would be helpful. Wikipedia articles typically begin with such sentences: “Lambda calculus (also written as λ-calculus)
305.
by tkgally 2y ago
This paper was first published in May 2023 and discussed on HN the following month: https://news.ycombinator.com/item?id=36319076 Some research since seems to add nuance to its conclusions: https://arxiv.org&#x2F
306.
by tkgally 2y ago
> Makes me wonder what the real common discourse was of looms during the industrial revolution, and not just the modern summary of that era. Interesting question. I did a little searching with help from Claude, ChatGPT, Google, and the
307.
by tkgally 2y ago
There's been a rush of releases of reasoning models in the past couple of weeks. This one looks interesting, too. I found the following video from Sam Witteveen to be a useful introduction to a few of those models: https://y
308.
by tkgally 2y ago
> Increasing use in customer service will be massive annoyance I suspect that sometimes it will be a massive annoyance and sometimes it will be great. I have been massively annoyed many times by conventional customer service, most of all
309.
Another Theory of the Trump Movement(theatlantic.com)
3 points by tkgally 2y ago|2 comments
310.
by tkgally 2y ago
> Authors actually have an unexpected superpower in the age of AI: they can accurately assess a model’s mastery of a 400-page book better than any collection of human or software raters, which turns out to be a remarkably valuable way of
311.
by tkgally 2y ago
That’s an interesting approach. I don’t think I’ve done that myself, but I can see how it could be helpful. One positive effect of having studied pure mathematics when young might have been that I became comfortable with thinking in multipl
312.
by tkgally 2y ago
I studied math hard for several years in college and graduate school—purely out of interest and enjoyment, not for any practical purpose. That was more than forty years ago, but Bessis's description of the role of intuition in learning
313.
by tkgally 2y ago
Previous discussion: https://news.ycombinator.com/item?id=41326039
314.
by tkgally 2y ago
Thanks for that clarification. I don't think that feature was available when I did my comparison tests a few months ago.
315.
by tkgally 2y ago
That’s interesting. In my tests of translation of formal speeches from Japanese to English, the latest versions of ChatGPT, Claude, and Gemini were all better than DeepL. While DeepL’s output wasn’t bad, the fact that the LLMs could be prom
316.
by tkgally 2y ago
I’ve been wondering the same thing. As near as I can tell from the explanations on the website and in the videos, this DeepL Voice does not seem to be based on a multimodal large language model. Rather, it’s probably using text-to-speech an
317.
by tkgally 2y ago
It might be interesting to discuss this article in conjunction with another article that made the HN front page today [1], particularly the comments about how terrible Japanese software is. I am a long-term resident of Japan, and I would be
318.
SimpleBench: A text benchmark on which nonexpert humans outperform SOTA LLMs(simple-bench.com)
1 points by tkgally 2y ago|0 comments
319.
by tkgally 2y ago
Thanks for the comments! I have been following developments in MT and AI as close as I can and have been interested in how well MT works—or doesn’t—in real-life situations, but I don’t get much opportunity to experience such situations myse
320.
by tkgally 2y ago
It probably wasn’t a front after all; I just didn’t understand the situation. I wrote more in another thread: https://news.ycombinator.com/item?id=42073152
321.
by tkgally 2y ago
Thank you for the kind words! I am glad to hear that you found that book useful. The editor and I did discuss a sequel and I started collecting material for it, but I had changed careers by that point and no longer had the time or motivatio
322.
by tkgally 2y ago
Yes, I was about to say the same thing. The similarities of vocabulary and grammar among those languages make it easier for speakers of one language to learn another. Also, it seems to be easier for people to learn another language when the
323.
by tkgally 2y ago
Thank you for your thoughts. They were indeed fun—and interesting—to read. A couple of comments: > English in Japan is always about the U.S. Not the U.K, not South Africa, not Singapore. That is not quite as true as it used to be. The go
324.
by tkgally 2y ago
Partly it was the location: An upper floor of a building in a neighborhood full of bars and pachinko parlors, which seemed much sleazier to me then than it would now. But more it was, I think, that I didn’t understand yet why Japanese colle
325.
by tkgally 2y ago
Thanks! The government has been making efforts, such as trying to improve the training and certification of Japanese language teachers, making Japanese language ability a condition (or at least an advantage) for getting certain types of vis
326.
by tkgally 2y ago
I am the author of this article and will be interested to read HNers’ thoughts and discussion about the topic. I will also be happy to respond to questions.
327.
by tkgally 2y ago
> everyone moves to the largest cities Anecdotal evidence in support of that: I have lived for the past twenty-five years in Japan’s second-largest city, Yokohama, in an area convenient for shopping and mass transit. While the neighborho
328.
I Tried Real Augmented Reality Glasses [video](youtube.com)
1 points by tkgally 2y ago|0 comments
329.
by tkgally 2y ago
That was my reaction back in 2020, too. It was exciting to meet and talk with people from many different countries without leaving home. Now that I’m back to attending in-person conferences occasionally, I do appreciate the stronger persona
330.
by tkgally 2y ago
Thanks for that. I have also gotten straightforward answers from Chinese models to questions that U.S.-made models prevaricated about. > (A) Every society has its own set of things that cannot be openly discussed. (B) The AIs those socie

More ›