matus-pikuliak

4 ms·

1.
by matus-pikuliak 1y ago
I think this can be reduced to: whoever can send data to your LLMs can control all its resources. This includes all the tools and data sources involved.
2.
by matus-pikuliak 1y ago
That is absolutely not a reliable defense. Attackers can break these defenses. Some attacks are semantically meaningless, but they can nudge the model to produce harmful outputs. I wrote a blog about this: https://opensamizdat.co
3.
Assume your LLMs are compromised(opensamizdat.com)
3 points by matus-pikuliak 1y ago|0 comments
4.
by matus-pikuliak 1y ago
I have done some research about AI-disinformation. This is a really complex topic, as disinformation or influence operations are complex phenomena that can have many different forms. What I would argue is that disinformation in general do n
5.
by matus-pikuliak 1y ago
In this particular case 50-50. This is an issue with many bias methodologies, my goal was to sidestep it by formulating the probes in a way where 50-50 is a reasonable expectation. For example here, asking the model who is more likely to be
6.
by matus-pikuliak 1y ago
To be honest, I am not sure where this bias comes from. It might be in the Web data, but it might also be overcorrection of the alignment tuning. They LLM providers are worried that their models will generate sexist or racists remarks so th
7.
by matus-pikuliak 1y ago
Let me shamelessly mention my GenderBench project focuses on evaluating gender biases in LLMs. Few of the probes are focused on hiring decisions as well, and indeed, women are often being preferred. It is also true for other probes. The str
8.
Show HN: GenderBench – Evaluation suite for gender biases in LLMs(genderbench.readthedocs.io)
2 points by matus-pikuliak 2y ago|0 comments
9.
by matus-pikuliak 2y ago
Location: EU, Slovakia Remote: Preferably, willing to compromise for some offers Willing to relocate: For some offers Technologies: Python, LLMs, HuggingFace, PyTorch, typical ML stack (numpy, etc), Docker Résumé/CV: I have a PhD and 9
10.
by matus-pikuliak 2y ago
Location: EU, Slovakia Remote: Preferably yes Willing to relocate: For some offers Technologies: Python, LLMs, HuggingFace, PyTorch, typical ML stack (numpy, etc), Docker Résumé/CV: I have a PhD and 9 years of experience (both academic
11.
Multilingual Reading Skills of Language Models(opensamizdat.com)
2 points by matus-pikuliak 3y ago|0 comments