My first language is Afrikaans, which is a somewhat niche language and hard to find teachers/conversation buddies outside South Africa. (I live in USA now)
I've been using Gemini to live chat in Afrikaans and do impromptu Afrikaans grammar lessons during my solo drives around town. It is phenomenal at speaking the language - like, it really shocks my family members when they hear it.
This is probably the most joy I get from any of my usages of LLMs/AIs. It's been really, really nice getting to speak my language regularly again. =)
So, I'm excited about this release and live chat getting better. I also hope the other frontier labs pick up niche languages like this as well so that I have more options.
I have a similar experience using Gemini for quick Catalan translations for iOS apps given enough context.
I once asked it to summarize The Hobbit in Catalan to explain it to my daughter before sleep. I was expecting a lot of mistakes as I see regularly if I ask anything in my native language when using GPT or Claude, but it was surprisingly good. I was going just to kind of skim ahead and retell it my own way, but ended up almost saying it verbatim because it was good already.
She loves Zelda so I asked it to explain the story of Breath of The Wild keeping the original names, and to make it fun, etc.. I was surprised again. I did retell some bits in my own style and taste but it is very convincing.
I haven't tried Catalan on newer models like GTP-6 Astra or Fable tho. We have all these benchmarks based on software development, and AGI, etc.. but it would be cool to have some language benchmarks for different communities.
As I work in english and use them in english, I wonder if using LLMs in a different language to code renders a different result as well. Like, if some of these benchmarks were made in other languages, would the result be similar.
> As I work in english and use them in english, I wonder if using LLMs in a different language to code renders a different result as well. Like, if some of these benchmarks were made in other languages, would the result be similar.
I am not a Chinese speaker but my understanding is that all of the models have substantially different behavior in Chinese, to the degree that it's kind of like a second model. Would be interested in hearing more if anyone has direct experience.
Fwiw, it's been very easy to nudge Qwen 27B/35B to think in Chinese. No need for <think> prefills like "思考：" ("Think:") or such.
Relatedly, when chatting with frontier models about science education content design, I've found it very helpful to mix in Chinese education terms. In English, for example, NGSS is such a massive attractor, discussing nearby topics often yields NGSS "slop". And "estimation" (educational) in the US means one (dysfunctional:) thing, which similarly distracts. Perhaps if AIs become increasingly multilingual, but remain weak at deep conceptual reasoning, it may be fruitful to have multilingual thesauruses, to use language-associated cultural conceptual differences as a way to convey conceptual nuances with which LLMs otherwise struggle?
This is an interesting post, but as a non-American it's hard to understand what's it's alluding to in some parts.
The one criticism of NGSS I found from a quick skim of <a href="https://en.wikipedia.org/wiki/Next_Generation_Science_Standards" rel="nofollow">https://en.wikipedia.org/wiki/Next_Generation_Science_Standa... is that it teaches evolution.
> What does "estimation" mean in the US?!
"Estimation" as a skill in US education, primary school on, is pervasively about point estimates - single numbers. Bounding, a range of safe/reasonable/likely/possible numbers, is very rare. In Fermi questions/problems also.
Even in such goodness as Sanjoy Mahajan's Art of Insight and Street-Fighting[1]... the only only "bound" mentioned is "Printed and bound in [USA]", and there is no "round".[2]
In contrast, IIUC, China is pervasively about "Da Gu" and "Xiao Gu", "round up" and "round down". Point estimates are secondary. Lower and upper bounds are apparently even a cultural concept: an individual's degree vs capability/fate; policy equity vs excellence; economic targets.
In education, my understanding is point estimates have much poorer conversational/collaborative dynamics than range estimates. Bounding lends itself to incremental collaboration. "Can anyone suggest another low or high bound?". Discussions around point estimates seem usually merely about the collection of individual estimates.
> NGSS
Ah, NGSS isn't bad particularly. I meant that discussion/mention of NGSS appears so very much in English training data, that AI chat on topics related to NGSS seem to draw regurgitations of NGSS. AI "slop" in the sense that "ignore NGSS, just reason from a superset of its underlying concepts"... isn't an LLM strength.
ERRATA: Those google book links now do have hits for "bound" and "round". Also "range" and "interval". The searches really didn't before, thus the reference links. Perhaps the absence should have failed sanity check, but given that one hit... partial failure of Google Books search wasn't something I recall noting previously. Oops.
Yes, and that brings in a shit load of slop to the problem space from the creationists.
Like how the word life will have a strong correlation with death, the word animate, while it also means alive will have a much less strong correlation with death.
Does European public dialog do much causal comparisons among states? Does somewhere else?
Or do you mean professionally... Instead of merely searching for "ways to do X", one might translate the query into a couple of languages, and translate the results, to get a broader culture sampling. Hmm... That seems something an AI-based search could integrate smoothly - cross cultural insights as a normal part of search results?
While there's a lot of diversity in policy and culture among US regions and states, contrasting them isn't leveraged much. Perhaps failing a "that's analysis, not news" threshold for being part of public dialog? So maybe it's not "non-English cultures" neglected as much as comparative discussion in general?
One "criminally underutilized" which really struck me, was Japan having a far more sensible and powerful COVID safety one-liner than the US managed - "Three C's" (Closed spaces, Crowded places, and Close-range contact - poorly-ventilated / crowd / conversation-contact - avoid overlapping all three) vs "6-foot rule".
jeanbza · · focus · HN ↗
I've been using Gemini to live chat in Afrikaans and do impromptu Afrikaans grammar lessons during my solo drives around town. It is phenomenal at speaking the language - like, it really shocks my family members when they hear it.
This is probably the most joy I get from any of my usages of LLMs/AIs. It's been really, really nice getting to speak my language regularly again. =)
So, I'm excited about this release and live chat getting better. I also hope the other frontier labs pick up niche languages like this as well so that I have more options.
LluisGerard · · focus · HN ↗
I once asked it to summarize The Hobbit in Catalan to explain it to my daughter before sleep. I was expecting a lot of mistakes as I see regularly if I ask anything in my native language when using GPT or Claude, but it was surprisingly good. I was going just to kind of skim ahead and retell it my own way, but ended up almost saying it verbatim because it was good already.
She loves Zelda so I asked it to explain the story of Breath of The Wild keeping the original names, and to make it fun, etc.. I was surprised again. I did retell some bits in my own style and taste but it is very convincing.
I haven't tried Catalan on newer models like GTP-6 Astra or Fable tho. We have all these benchmarks based on software development, and AGI, etc.. but it would be cool to have some language benchmarks for different communities.
As I work in english and use them in english, I wonder if using LLMs in a different language to code renders a different result as well. Like, if some of these benchmarks were made in other languages, would the result be similar.
jaggederest · · focus · HN ↗
I am not a Chinese speaker but my understanding is that all of the models have substantially different behavior in Chinese, to the degree that it's kind of like a second model. Would be interested in hearing more if anyone has direct experience.
mncharity · · focus · HN ↗
Relatedly, when chatting with frontier models about science education content design, I've found it very helpful to mix in Chinese education terms. In English, for example, NGSS is such a massive attractor, discussing nearby topics often yields NGSS "slop". And "estimation" (educational) in the US means one (dysfunctional:) thing, which similarly distracts. Perhaps if AIs become increasingly multilingual, but remain weak at deep conceptual reasoning, it may be fruitful to have multilingual thesauruses, to use language-associated cultural conceptual differences as a way to convey conceptual nuances with which LLMs otherwise struggle?
eru · · focus · HN ↗
For that particular attractor, even German or Spanish or so might help avoid it? Or perhaps even just using British English?
mncharity · · focus · HN ↗
normie3000 · · focus · HN ↗
The one criticism of NGSS I found from a quick skim of <a href="https://en.wikipedia.org/wiki/Next_Generation_Science_Standards" rel="nofollow">https://en.wikipedia.org/wiki/Next_Generation_Science_Standa... is that it teaches evolution.
What does "estimation" mean in the US?!
mncharity · · focus · HN ↗
"Estimation" as a skill in US education, primary school on, is pervasively about point estimates - single numbers. Bounding, a range of safe/reasonable/likely/possible numbers, is very rare. In Fermi questions/problems also.
Even in such goodness as Sanjoy Mahajan's Art of Insight and Street-Fighting[1]... the only only "bound" mentioned is "Printed and bound in [USA]", and there is no "round".[2]
In contrast, IIUC, China is pervasively about "Da Gu" and "Xiao Gu", "round up" and "round down". Point estimates are secondary. Lower and upper bounds are apparently even a cultural concept: an individual's degree vs capability/fate; policy equity vs excellence; economic targets.
In education, my understanding is point estimates have much poorer conversational/collaborative dynamics than range estimates. Bounding lends itself to incremental collaboration. "Can anyone suggest another low or high bound?". Discussions around point estimates seem usually merely about the collection of individual estimates.
> NGSS
Ah, NGSS isn't bad particularly. I meant that discussion/mention of NGSS appears so very much in English training data, that AI chat on topics related to NGSS seem to draw regurgitations of NGSS. AI "slop" in the sense that "ignore NGSS, just reason from a superset of its underlying concepts"... isn't an LLM strength.
> hard to understand
Sorry - Thanks for asking!
[1] open access: <a href="https://direct.mit.edu/books/oa-monograph/5345/The-Art-of-Insight-in-Science-and" rel="nofollow">https://direct.mit.edu/books/oa-monograph/5345/The-Art-of-In... <a href="https://direct.mit.edu/books/oa-monograph/5339/Street-FightingMathematicsThe-Art-of-Educated" rel="nofollow">https://direct.mit.edu/books/oa-monograph/5339/Street-Fighti... [2] search "bound": <a href="https://www.google.com/books/edition/The_Art_of_Insight_in_Science_and_Engine/xRgeBQAAQBAJ?hl=en&gbpv=1&bsq=%20bound" rel="nofollow">https://www.google.com/books/edition/The_Art_of_Insight_in_S... <a href="https://www.google.com/books/edition/Street_Fighting_Mathematics/VrkZN0T0GaUC?hl=en&gbpv=1&bsq=bound" rel="nofollow">https://www.google.com/books/edition/Street_Fighting_Mathema...
mncharity · · focus · HN ↗
pixl97 · · focus · HN ↗
Yes, and that brings in a shit load of slop to the problem space from the creationists.
Like how the word life will have a strong correlation with death, the word animate, while it also means alive will have a much less strong correlation with death.
tomaskafka · · focus · HN ↗
mncharity · · focus · HN ↗
Or do you mean professionally... Instead of merely searching for "ways to do X", one might translate the query into a couple of languages, and translate the results, to get a broader culture sampling. Hmm... That seems something an AI-based search could integrate smoothly - cross cultural insights as a normal part of search results?
While there's a lot of diversity in policy and culture among US regions and states, contrasting them isn't leveraged much. Perhaps failing a "that's analysis, not news" threshold for being part of public dialog? So maybe it's not "non-English cultures" neglected as much as comparative discussion in general?
One "criminally underutilized" which really struck me, was Japan having a far more sensible and powerful COVID safety one-liner than the US managed - "Three C's" (Closed spaces, Crowded places, and Close-range contact - poorly-ventilated / crowd / conversation-contact - avoid overlapping all three) vs "6-foot rule".