Hacker News Reader

Comment by icapybara | Hacker News Reader

Why wouldn't you want an LLM for a language learning tool? Language is one of things I would trust an LLM completely on. Have you ever seen ChatGPT make an English mistake?

View on HN

Grammarly is all in on AI and recently started recommended splitting "wasn't" and added the contraction to the word it modified. Example: "truly wasn't" becomes "was trulyn't"

https://imgur.com/a/RQZ2wXA

Hm ... I wonder, is Grammarly also responsible for the flood of contraction of lexical "have" the last few years? It's standard in British English, but outside of poetry it is proscribed in almost all other dialects (which only permit contraction of auxiliary "have").

Even in British I'm not sure how widely they actually use it - do they say "I've a car" and "I haven't a car"?

"they" say "I haven't got a car".

Contractions are common in Australian English to, though becoming less so due to the influence of US English.

In my experience "I've a car" is much more common than "I haven't a car" (I've never heard the latter construct used, but regularly hear the former in casual speech). "I haven't got a car" or "I've no car" would be relatively common though.

This is what peak innovation looks like

I don't think an LLM would recommend an edit like that.

Has to be a bug in their rule-based system?

Gemini: "Was trulyn't" is a contraction that follows the rules of forming contractions, but it is not a widely used or accepted form in standard English. It is considered grammatically correct in a technical sense, but it's not common usage and can sound awkward or incorrect to native speakers.

I wonder how much memes like whomst'd might skew the training set.

Yeah, I agree. An open-source LLM-based grammar checker with a user interface similar to Grammarly is probably what I'm looking for. It doesn't need to be perfect (none of the options are); it just needs to help me become a better writer by pointing out issues in my text. I can ignore the false positives, and as long as it helps improve my text, I don't mind if it doesn't catch every single issue.

Using an LLM would also help make it multilingual. Both Grammarly and Harper only support English and will likely never support more than a few dozen very popular languages. LLMs could help cover a much wider range of languages.

I tried to use one LLM based tool to rewrite sentence in more official corporate form, and it rewrote something like "we are having issues with xyz" into "please provide more information and I'll do my best to help".

LLMs are trained so hard to be helpful that it's really hard to contain them into other tasks

uh. yes? it's far from uncommon, and sometimes it's ludicrously wrong. Grammarly has been getting quite a lot of meme-content lately showing stuff like that.

it is of course mostly very good at it, but it's very far from "trustworthy", and it tends to mirror mistakes you make.

Do you have any examples? The only time I noticed an LLM make a language mistake was when using a quantized model (gemma) with my native language (so much smaller training data pool).

Not GP, but I've definitely seen cutting edge LLMs make language mistakes. The most head scratching one I've seen in the past few weeks is when Gemini Pro decided to use <em> and </em> tags to emphasize something that was not code.

Because this "language learning tool" will be dominantly used to avoid actually learning the language.

Reply on news.ycombinator.com