First, that's UCSD, not UC Davis. They mention UC Davis in the report, but the 8.5% refers to UCSD.
Secondly, mastering high school math is genuinely difficult these days. I'm a math major, I've made it through my calc courses and differential equations, but I found Algebra 2 legitimately hard. Logarithms and Trigonometric functions are counterintuitive, and not everybody is at their peak ability to buckle down and grind through things when they're struggling at age 17.
And lastly, this is pretty obviously at least in part a knock-on affect of covid, hence the extremely recent major spike. I'm not sure it's worth generalizing from "UC San Diego Students admitted in the last couple of years are struggling with high school math (because they were in high school during lockdown)" to "We shouldn't try paying mechanics more because everyone's bad at math"
While Math 2 was designed in 2016 to remediate missing high school math knowledge, now most students had knowledge gaps that went back much further, to middle and even elementary school. To address the large number of underprepared students, the Mathematics Department redesigned Math 2 for Fall 2024 to focus entirely on elementary and middle school Common Core math subjects (grades 1-8), and introduced a new course, Math 3B, so as to cover missing high-school common core math subjects (Algebra I, Geometry, Algebra II or Math I, II, III; grades 9-11).
So actually I was wrong. I should have said:
- 11.8% of UCSD freshmen haven't mastered high school math
- 8.5% haven't mastered middle school math
These folks may have had some disruption during the last year of middle school, and the first year of high school. But does that fully explain why they haven't mastered middle school math or, in some cases, elementary school math?
The comment to which I responded quibbled with mhb saying "We used to be able to assume that high school and college graduates could do elementary school math."
It's clear from the report I linked that we cannot assume that high school graduates can do elementary school math.
> It's clear from the report I linked that we cannot assume that high school graduates can do elementary school math.
Well, I wouldn't necessarily assume that 100% of anyone with a degree has mastered what the degree is for. So to me the takeaway is that ~90% have mastered the math. And so in terms of the original comment, not necessarily do we need them all to go to undergraduate.
For this use case, why not use Whisper to transcribe the audio, and then an LLM to do a second step (summarization or answering questions or whatever)?
I would much rather learners be directed at a proper resource for doing something than trying to include all the info locally which inevitably will get out of date and become incomplete.
Is a search term really a proper resource? A chosen installation guide, preferably an official one with a stable URL (and available in the language of instruction) would be better IMHO. When the link goes dead, the learner could search based on the link title anyway.
The materials are available for non-chinese learners. Although from the overview you can know that the first attempt of this project is to train skillful Chinese students for industry/academic research。
In my city, the regulations specify a maximum # kids per adult. So if you were to devise a way to supervise more children per adult, using technology, you would still have to hire the same number of adults.
The regulations specify that teachers must have completed a certain number of units of a specific type of education. If you create an AI Assistant that means you can hire people with less training and have the same quality, then ... you cannot.
The regulations regulate inputs rather than outputs.
Well I think the regulations regulate outputs as well (if a child dies or is injured in daycare, there are regulations to handle that). The issue is that people aren't happy with settling for reactive punishments when something actually goes wrong.
In London, the M25 (outer ring road) has variable speed limits. Perhaps it's to improve throughout on the M25, or to increase the capacity of the M25 to absorb traffic from nearby roads.
The article addresses this. It explains that if large portion of your job can be automated (and hence be done at little marginal cost) then the remaining parts of your job become more valuable.
The article does not accurately reflect even the abstract, so I would not count on it to accurately reflect the full paper.
From the article (emphasis mine):
the researchers analyzed shifting dynamics of over 11,000 families with kids *for a decade*
From the paper abstract (emphasis mine):
The present study examined the longitudinal, bidirectional associations between early adolescents’ electronic media use and family conflict *across 3 years*
- the README is extremely detailed and clear: all the commands are explained with examples and the why to use each one
- you're using Anki Connect to edit decks in-place, instead of trying to edit or generate an apkg file. This simplifies things and avoids issues such as needing to create custom note types or avoiding creating two note types with the same field
When my son and I have discussed a topic in response to a question, ideally I would evaluate whether there's something he should remember forever and, if so, I would create one or more Anki notes for that piece of knowledge. But right now it's too much effort, unless I'm at my desk. Even then, I need to copy and paste card fields from a chat interface into the Anki UI. That means I rarely do it.
reply