Sander Schulhoff
speaker
947 appearances
1 recordings
1 series
first heard Jun 2025
last heard Jun 2025
Sander Schulhoff’s voice in public audio — every appearance, attributed to the second.
Trend
recordings per month · last 12 monthsNo recordings in the last 12 months.Older appearances are listed below; set an alert to hear about the next one.
Appearances
It's like context, giving you more context.
Giving more context.
Uh exactly.
Uh and so that's why that one might work, might have worked, and for the kind of threats and promises
I've seen explanations of like, oh, the the AI was trained with like reinforcement learning, so it it knows to learn from rewards and punishments, which
Like is is true in a rather pure mathematical sense, but
I just I don't feel like it works quite like that with the prompting.
Like that's not how the training is done.
Like during training, it's not told, hey, like, do a good job on this and you'll get paid and then like that that's just not how training is done.
Uh and so that's why
Uh I don't think that's a great explanation.
decomposition is another really, really effective technique.
And for most of the techniques that I will discuss, you can use them in either the conversational or the product focused setting.
Uh and so for decomposition, the core idea is that there's some task, some task in your prompt that you want the model to do.
Uh and
If you just ask it that task straight up, it might kind of struggle with it.
So instead you give it this task and you say, hey, don't answer this.
Before answering it, tell me
What are some sub problems that would need to be solved first?
Uh and then it gives you a list of subproblems.
Showing 201–220 of 947 · page 11 of 48
← Previous
Next →