EveYoung · HackerTrans

3 years agodiscuss

To me, the fact that you can configure your "GF" sounds very creepy. No real person is always happy, positive, or understanding. Especially for people with poor social skills, this sounds like it would just exacerbate their problems and make it more difficult for them to connect with humans.

EveYoung

3 years agodiscuss

Would you run the LLM extractor across every page? Especially for larger scale projects, such as scraping entire product catalogues, this sounds very expensive. Maybe you could use the AI to generate selectors from examples that can then be applied to all other pages of the same structure?

EveYoung

3 years agodiscuss

The problem is that for longer journeys 4G/5G coverage is often very spotty.

EveYoung

3 years agodiscuss

Are there even similar solutions available commercially? Even at a size of JPM, I highly doubt that they would develop such a complex system in-house. Also, for any high-skill job, I feel such an "activity metric" is utterly useless and not correlated with value contributions.

EveYoung

3 years agodiscuss

I think it's unrealistic to expect a multi-national government funded organisation to offer a commodity service at a competitive price. With all the conflicting national interests and bureaucracy that come with this funding model, how should they compete with much leaner, commercial organisations that are only responsible to their shareholders? But does it really matter? If the main objective of the Ariane Group is to offer Europe independence by providing access to space, I think that’s worth a premium.

EveYoung

3 years agodiscuss

In that case, you could even enable the sharing of session cookies as well.

EveYoung

3 years agodiscuss

If older techniques work for you, there is of course no reason to switch to LLMs besides general curiousity or to explore what's possible already. That said, in my case I was enable to generate much more engaging text summaries of tabular data using a Llama derivative.

EveYoung

3 years agodiscuss

In my experience, the threshold to be useful is much lower than GPT-3.5. These smaller models can "easily" be finetuned to achieve a comparable performance on a specific task. For example, I've achieved promising results for data summarisiation and image captioning (BLIP2-based) using Alpaca.

Also, server/hardware costs are still a limiting factor for running and finetuning the larger 33/65B Llama models. Especially, if they can only be used for personal toy projects.

EveYoung

3 years agodiscuss

In the long-term, I doubt that OpenAI will have a monopoly on LLMs. We're still very early in the development, and I don't see why other organisations, won't be able to catch up. The instruction-tuned LLaMa versions already look very promising, even with rather small & low-quality data sets. Compared to other technologies, the investment-requirements for LLMs still seem to be relativly low and something a number of companies or governments are able to finance.

EveYoung

3 years agodiscuss

Do you know if they adapt the shows to the countries they're in? I'm wondering how universally understood clowns are.

EveYoung

4 years agodiscuss

Why not have both? Earlier in my career, I was very invested in my job as well and didn't mind working 12h+ as I was learning a lot. And I think this mindset really helps to progress quickly. But at some point, it just became unsustainable. I still love my job but also my personal life outside of work.

EveYoung

4 years agodiscuss

I think the problem for most people is that we simply don't have many opportunities for these type of interactions at a certain point in life. Also, why do you need to have deep conversations before going out on a date? If there is physical attraction and some shared interests, what's the harm of meeting for some drinks?

EveYoung

4 years agodiscuss

For me, it always comes down to genuine self-esteem. Someone's fashion style, physic, or facial symmetry really doesn't matter if they have no confidence in themselves.

comments