logoalt Hacker News

hedoratoday at 7:26 PM2 repliesview on HN

The difference between fifth and sixth grade reading levels doesn’t have much to do with sounding out words.

Reading text aloud to the illiterate group is not sufficient to let them understand the material.

Arguably, an LLM could rewrite the content to target illiterate audiences.

chatgpt 4o mini did better with your comment than I’d guessed it would:

> With new tools that change speech to text and text to speech, many people might not need to read or write as much. If someone has to make an important paper, like a contract or a will, they will probably ask someone who knows how to do that.

> It's not clear if people who work in STEM (like science and math) can skip reading and writing. In STEM, they use many pictures, charts, and drawings to share ideas. While these don’t always use words, they often have things like math symbols or simple sentences to help explain.

I feel the sentence structures it chose are a bit too complicated for K-3rd grade level, which is what it should have targeted. Maybe some prompt engineering could get it to simplify further.


Replies

Merrilltoday at 7:54 PM

Very interesting that it did so well. Although it did leave out the point that reading and interpreting documents could also be done by an expert in the first paragraph as well as the idea that not just words but language are used in technical communications by the various symbolic systems.

I think that sophisticated verbal communications can be learned by verbal means, and that the reading of literature is not essential. Non-literate cultures have maintained traditional folk songs, storytelling and epic poetry.

Corporate managers and salespeople are often highly verbal, but not necessarily highly literate. Consider how the written Response to an Request For Proposals is not enough for important opportunities, but must be simplified to a set of slides delivered by a silver-tonged senior salesperson. This provides a better match to the input characteristics of the customer's decision makers.

show 1 reply
selimthegrimtoday at 7:37 PM

Good rebuses are hard, it should be interesting to see whether VLMs can be trained to judge them