TomasBM

12 ms·

91.
by TomasBM 11mo ago
Looks great, and I definitely like the approach, but I wonder who would be the ideal user here. Regional pricing is not that difficult of a problem, unless you're targeting particular customer bases and want to introduce discounts. So,
92.
by TomasBM 11mo ago
I've seen arguments that opening up fresh material makes it easy for less honest institutions to plagiarize your work. I've even heard professors say they don't want to share their slides or record their lectures, because it&
93.
by TomasBM 11mo ago
It's fair to worry about the geopolitics of borders and minority groups, but this does not have much to do with a company placing its headquarters in a middle country where maybe (former) members of a controversial state could also wor
94.
by TomasBM 11mo ago
You're objectively wrong. "Sanctioned individuals" from Russia would have a hard time working in Belgrade, since it is under a lot of pressure and scrutiny from "the West". These individuals would have an easier tim
95.
by TomasBM 11mo ago
Yeah, this resonates with me. As much as I dislike not having a good mental model of all the code that does things for me, ultimately, I have to concede the battle to get things done. This is not that different from importing packages that
96.
by TomasBM 1y ago
Yes, and it's not a sound argument, IMO. "Grasping concepts" is not the only learning goal in schools or universities. Many classes - including within STEM programmes - want to teach students about writing, argumentation, res
97.
by TomasBM 1y ago
If you're asking this seriously: We need to grade people because that's the best way we have to determine (for one or more subjects) who's: 1. capable enough, so that we can promote them to the next stage; 2. improving or
98.
by TomasBM 1y ago
I see what you did there. I think we'll still need ways to detect copy-pasted zero-shot content that's generated by LLMs, for the same reasons that teachers needed ways to detect plagiarism. Kids, students, and interns [1] "c
99.
by TomasBM 1y ago
You might want to look into knowledge graphs (KGs), graph databases, ontologies, and similar. I personally and professionally used these to do some cool things, like run audits across different systems simultaneously. Common stack would inc
100.
by TomasBM 1y ago
I'm not a 'native' programmer, but even I can agree with that. As it became easier to abstract away from pure assembly, people did just that. Some ideas stuck, some converged, and the ones with the most tenacious, adaptable a
101.
by TomasBM 1y ago
> There's no other relation than "causes". Looking at their Neo4j graph, they also retain the provenance of the causal relation in "claimedIn" relations between the reified triple of each cause-effect pa
102.
by TomasBM 1y ago
I think we're referring to different debts . Your original comment said: > at some point in the future the rich have to pay that work back with something of equal value [e.g. food, gadgets, etc.]. And this is true in the ab
103.
by TomasBM 1y ago
So, maternity/paternity leave in line with European systems? Or did you have something else in mind?
104.
by TomasBM 1y ago
I don't know what/who the "security-complex handlers" are, but it's quite useful to start with the supposedly biggest repository of human knowledge. Maybe you realize the causal relations you tease out are low-quali
105.
by TomasBM 1y ago
You can think even simpler. As one individual, you don't really owe anything to anyone. The only time you owe something is in social terms, when you borrow it in your name, or promise reward for work. And even then, people try to get
106.
by TomasBM 1y ago
Sounds like real-life Factorio.
107.
by TomasBM 1y ago
I'd rather say LLMs are a lossy encyclopedia + other things. The other things part obviously does a lot of work here, but if we strip it away, we can claim that the remaining subset of the underlying network encodes true information
108.
by TomasBM 1y ago
I'm actively working with ontologies (disclaimer: as a researcher), and yours is the top comment, so I'll try to make some counterclaims here. No relation to this work tho. > Ontologies and all that have been tried and have alw
109.
by TomasBM 1y ago
Cyc is hardly [1] mentioned in modern work under the knowledge representation and reasoning umbrella, because most [2] of it was/is unavailable or unknown to most researchers. It's hard to build on something that's primarily
110.
by TomasBM 1y ago
"It's not that I have something to hide. It's that I have nothing I want you to see." A great quote from an otherwise OK movie ("Anon").
111.
by TomasBM 1y ago
Slightly tangential: A lot of these issues are philosophical in origin, because we don't have priors to study. But just because, for example, advanced nanotechnology doesn't exist yet, that doesn't mean we can't imagine
112.
by TomasBM 1y ago
Slightly tangential: A lot of these issues are philosophical in origin, because we don't have priors to study. But just because, for example, advanced nanotechnology doesn't exist yet, that doesn't mean we can't imagine
113.
by TomasBM 1y ago
They're not really comparable. AI is not only a general-purpose technology (such as, e.g., electricity or computers), but also the only one that can self-improve. Also, its potential for diffusion is much bigger, because unlike for ele
114.
by TomasBM 2y ago
Perhaps it's not just about the 'shelf life' of a book, but also the language and style they use. The more archaic the language, and the more distant the style that the author's use, the harder it is for me to focus on t
115.
by TomasBM 2y ago
Outside the US (e.g., Europe), advertising bans for tobacco and liquor can work quite well. Sure, there will always be bad cases and loopholes - even bans on murder don't work 100% - but there's a reason "bans" are still
116.
by TomasBM 2y ago
Although I can appreciate your point about having some 'innate' desire for an activity like coding, I think this desire is just one of many factors in choice of work. My own anecdotal experience is that because of several factors,
117.
by TomasBM 2y ago
There's no need to believe in an immaterial soul to think that a copy is a different 'you'. It is enough to understand that having an independent "command & control" structure (for lack of a better word; C&C
118.
by TomasBM 2y ago
I believe you're right, even though I don't have any evidence except for my own experience. This issue becomes very clear when you see how many ways there are to express a simple concept like linear regression. I've had the c
119.
by TomasBM 2y ago
Depends on what you're writing, I guess. If you're writing a scientific article with the results of your research, your audience (i.e., reviewers and other readers, usually scientists) will expect you to have evidence for your cla
120.
by TomasBM 2y ago
I'm working on ways to allow developers and deployers of LLMs to express how and why their overall system is compliant and adversarially robust, and what to do when that's not the case. Specifically, my team and I are making assur

More ›