Oranguru

You are too deeply mired in propaganda for any meaningful debate; you should read more neutral, international news sources to get a clearer picture. This was all about oil; not drugs, not terrorism, not communism. One less criminal is nothing for such a corruputed state. The US has simply gained yet another vassal state. Nothing else has changed. Venezuelan people are in the same shit as before.

AI has not yet reached the point where it can convincingly generate long presentations with the precision required for complex, technical academic topics. It will eventually get there, at which point an additional layer of proof of work will need to be introduced. This is an adversarial race.

I think this is an interesting and effective solution, at least for now. Similarly, graduate and master's theses should focus more on the presentation and on eliciting knowledge from the students through critical, thorough questioning than on the tangible outcome of the project, which can easily and bindly be obtained with AI these days.

Nevertheless, Venezuela remains in the same situation as before. Aside from Maduro, the corrupt and criminal government remains intact, the head of state handpicked by the US president, no public elections in sight. There has been no political change whatsoever apart from the takeover of oil fields.

> We're all more likely to get what we want. Ideally peace and prosperity.

As sovereign citizens of their own country, Ukrainians have the right to decide under which flag they want to collectively pursue peace and prosperity. I believe they have made that choice clear through democratic means, something Russians have not experienced in a long time. If Russia is allowed to take Ukraine, it will move on to Poland next, and then to claim the remaining former Soviet states. A line must be drawn somewhere.

This is different now. US labs and companies are not releasing frontier-level models openly (specially those capable of assisting cyber intelligence work), but commercializing them instead. Thus, any ban would not be symmetrical to begin with, and that is precisely what maintains the balance.

They didn't say that; they simply said this is not the way of democracy. Legality has nothing to do with it.

But this isn't a story, literature, or a fancy piece of art; it's merely a technical blog post that discloses a security vulnerability. Here, the writing serves only as a vehicle to convey a message. Once you've received it, its purpose has been fulfilled. I would agree with you if the writing were an important part of the message, but here it is not. Not everybody can write well, and this guy clearly had something to tell, and that is what matters.

Prompt injection is still a possibility, so while it improves the security posture, not by much.

What "other platforms" are you talking about? Just an Android app would suffice. It's not a huge deal for a company worth trillions, especially if the features are already there and they're just blocking non-Apple products. If they deliberately do that, it makes you think they don't really care about their customers and are more interested in locking people into their ecosystem.

You can absolutely sync your vault without a paid subscription. Simply save it within your OneDrive or Google Drive folder. Alternatively, you could use Syncthing if you prefer a self-hosted solution.

However, these rights should be guaranteed to a company operating in the USA and strictly adhering to US law. Of course, if the law is (arbitrarily) changed to make this illegal due to the Chinese government's stake, then it could be forced to shut down, but that would be inconsistent with the constitution.

Because both of these are based on Chromium (the open-source version of Chrome).

This approach is valuable because it abstracts away certain complexities for the user, allowing them to focus on the code itself. I found it especially beneficial for users who are not willing to learn functional languages or parallelize code in imperative languages. HPC specialists might not be the current target audience, and code generation can always be improved over time, and I trust based on the dev comments that it will be.

You can easily fix this using a grammar constraint with llama.cpp. Add this to the command: --grammar "root ::= [^一-鿿ぁ-ゟァ-ヿ가-힣]*"

This will ban Chinese characters from the sampling process. Works for Yi and Qwen models.

You can access GPUs within containers using CDI (Container Device Interface): https://docs.nvidia.com/datacenter/cloud-native/container-to... No additional tools (e.g., nvidia-ctk) are needed. Docker has recently added support for CDI in version 25.0.

Very interesting. Thanks for the references. Have you released the code or pre-trained models yet or do you plan to do so at some point?

Thank you for your work. I have been having trouble achieving the desired level of formality in the generated text. When I ask for slightly formal content, the result tends to be too formal. However, when I ask the model to reduce the formality or use a semi-formal tone, the text becomes too informal. This will allow me to exercise more control over the style of the model's output and stop constantly battling with it.

This comment does nothing more than repeat the information provided in the title and in a very verbose style. I don't understand the tendency for GPT generated comments that is lately taking over HN and Twitter threads. They are insipid, add absolutely nothing of value to the discussion, and are generally a waste of time to read.

It's about minimising friction in our tasks and reducing any unnecessary obstacles. Even seemingly minor actions that only take a few seconds can build up over time and generate a sense of frustration, especially when they become frequent. Personally, I have found that dealing with this kind of friction can erode my overall productivity, as I unconsciously shy away from these tedious tasks that involve manual repetition, no matter how small. That's why many of us choose to invest a little time in coming up with these small automations that free us from the clutches of monotonous and repetitive tasks, allowing us to focus on the core of our work.

Great job!

In the future, would you add support for local LLMs, such as LLaMa?

Yes, check out: https://huggingface.co/chat/

You can easily opt out of the data sharing.

Useless for what? Are you comparing the base model with chat-tuned models?

Chat-tuned derivatives of LLaMa 2 are already appearing. Given that the base LLaMa 2 model is more efficient than LLaMa 1, it is reasonable to expect that these more refined chat-tuned versions of the chat-tuned versions will outperform the ones you mention.

"Compared to GPT2 it’s on par" Any benchmarks or evidence tu support this claim? IF you try to find them, official benchmarks will tell you that this is not true. Even the smallest LLaMa model (7B) is far ahead of GPT2, like an order of magnitude better in perplexity.

This is also a requirement for the research conducted under a project funded by the european comission.

Take a look at: https://huggingface.co/blog/trl-peft

Yes, there are. For example: https://laion.ai/blog/open-flamingo/

I must remind you that large language models are not designed to perform arithmetic calculations nor they have been trained to do so. They are trained to recognize patterns in large amounts of text data and generate responses based on that learned information. While they may not be able to perform some specific tasks, they can still provide useful information and insights in a wide range of applications. Judging their quality of *language* models because of their inability to do basic math is completely unfair.

Hallucinating? you mean overfitting, right?

For STT, take a look at Wenet: https://github.com/wenet-e2e/wenet

They provide support for running in a Raspberry Pi and it runs in real-time. I have tried the desktop version and the quality is good enough when the audio is clean.

This technical report from the European Commission was very useful in the research I conducted on this topic earlier this year.

https://publications.jrc.ec.europa.eu/repository/handle/JRC1...