rzmmm

9 ms·

91.
by rzmmm 7mo ago
In academia there has been a widespread practice to simply include a sentence about how AI has used in articles. It's simple and it works well.
92.
by rzmmm 7mo ago
Yes. Can someone tell me why even HN has bots. For selling upvotes to advertisement purposes?
93.
by rzmmm 7mo ago
Someone compared LLM in 2020s to GUI in 80s and 90s. Graphical interfaces didn't replace text interface, but it just became additional to it.
94.
by rzmmm 7mo ago
I'm a bit confused about these replies. The user was talking about posting AI summaries in HN comments. I suggested that posting an URL may be better choice.
95.
by rzmmm 7mo ago
Perplexity supports sharing URL to the thread. I think it's quite natural to link AI summaries like that.
96.
by rzmmm 7mo ago
Yeah can't find much information either. I would like to see at least some proof. Either via Mckinsey or from the security team.
97.
by rzmmm 7mo ago
Bingo. I can see this is a possible future, and probably desirable scenario for anyone with preference for free software.
98.
by rzmmm 8mo ago
Looks like it supports alternative proof of employment. They don't require disclosing identity as long as they are convinced you work for these companies.
99.
by rzmmm 8mo ago
I don't have solutions but I have similar experiences about this. It's probably a difficult problem since there are so many different queries and differences in the geospatial data.
100.
by rzmmm 8mo ago
I think "understand this directory deeply" just gives more focus for the instruction. So it's like "burn more tokens for this phase than you normally would".
101.
by rzmmm 8mo ago
It looks like typical "memorization" in image generation models. The author likely just prompted the image. The model makers attempt to add guardrails to prevent this but it's not perfect. It seems a lot of large AI models ba
102.
by rzmmm 8mo ago
That's interesting. Here is more information https://sqlite.org/th3.html The roots of SQLite are in defence industry projects of US Navy and General Dynamics. Seems like TH3 might be of interest for these sort of users
103.
by rzmmm 8mo ago
The bar is higher for AI in most cases.
104.
by rzmmm 8mo ago
I think it's technically possible to achieve determinism with LLM output. The LLM makers typically make them non-deterministic by default but it's not inherent to them.
105.
by rzmmm 8mo ago
Genetics may predispose for nicotine addiction, obesity, alcoholism, etc. This is intended design of genetic studies which look at multiple genes, like twin studies, GWAS studies, etc. I think the studies which find a single gene variant wh
106.
by rzmmm 8mo ago
Geneticist would say it's intended design of twin studies. Your example doesn't show AIDS is a genetic defect, but that genetics may predispose/protect against it.
107.
by rzmmm 8mo ago
Environmental effects are not necessarily modifiable. It includes randomness, background radiation, unknown risk factors, anything which is not genetic.
108.
by rzmmm 9mo ago
I feel that this is too costly for that kind of usage. Probably quote different architecture is needed for robotics.
109.
by rzmmm 9mo ago
There is an interesting study in older adults who were former elite athletes. Knee osteoarthritis (knee joint wear-and-tear) was best predicted by high body mass and knee injuries as young adults. https://pubmed.ncbi.nlm.nih.gov&
110.
by rzmmm 9mo ago
You could try replacing "if unsure..." with "if even slightly unsure..." or so. The verbosity and anthropomorphism is unnecessary.
111.
by rzmmm 9mo ago
He is a familiar blogger for HN readers, has been for a long time. While I agree the posts are nowadays a bit repetitive, he has also very interesting non-AI content. Some people probably upvote because they like the author, not necessarily
112.
by rzmmm 9mo ago
They can probably use most of SQLite test suite.
113.
by rzmmm 9mo ago
Very cool project. Edward Tufte presented decades ago that great visualizations maximizes data-ink ratio. This is what he ment ;)
114.
by rzmmm 9mo ago
My impression is that AI models need large amounts of quality training data. "Data contamination", i.e. AI output in the training data set has been a problem for years.
115.
by rzmmm 9mo ago
I'm not sure OP is looking for evidence like this. There are many optimistic articles from people or organizations who are selling AI products, AI courses, or AI newsletters.
116.
by rzmmm 9mo ago
The model has multiple layers of mechanisms to prevent carbon copy output of the training data.
117.
by rzmmm 9mo ago
Someone presented a hypothetical scenario: What if a hacker would write a virus, which breached a totally unprotected database after the hacker has passed away. It's clear that the therapy provider is at least partially responsible.
118.
by rzmmm 9mo ago
I believe the industry has largely accepted that prompt injection is inherent part of LLM tech.
119.
by rzmmm 9mo ago
There are many open-source toy browser implementations available, so this seems quite likely.
120.
by rzmmm 9mo ago
Maybe influential would be a better word choice. Firefox has 100M+ users.

More ›