←
David Holmes497 posts
David Holmes avatar

David Holmes

@DavidHolmes

Long-time blogger & digital curator | sound engineer & musician | Internet music aficionado. Specialized lists available.

◷ Joined March 2007▧ 2 Photos and videos
David Holmes avatar
David Holmes@DavidHolmes·19 Dec 2023

With the talent of artists and new tools, imaginary worlds and characters are growing more enthralling and vivid.

Chikai@lifeofc18 Dec 2023

A new cinematic AI piece. Continuing with the @Magnific_AI + @midjourney + @runwayml combination, but in the setting of a museum. It's an introspection on who is watching whom in the art world.👀 "Panopticon - Museum"

1:09
Chris Stokel-Walker avatar
⟳ David Holmes retweeted
Chris Stokel-Walker@stokel·14 Dec 2023

Grok, xAI's LLM chatbot, started referring to OpenAI in responses over the weekend. It could be the first warning sign of a significant collapse in our collective knowledge, experts fear. By me for @FastCompany fastcompany.com/90998360/gro…

Andrej Karpathy avatar
⟳ David Holmes retweeted
Andrej Karpathy@karpathy·9 Dec 2023

# On the "hallucination problem" I always struggle a bit with I'm asked about the "hallucination problem" in LLMs. Because, in some sense, hallucination is all LLMs do. They are dream machines. We direct their dreams with prompts. The prompts start the dream, and based on the LLM's hazy recollection of its training documents, most of the time the result goes someplace useful. It's only when the dreams go into deemed factually incorrect territory that we label it a "hallucination". It looks like a bug, but it's just the LLM doing what it always does. At the other end of the extreme consider a search engine. It takes the prompt and just returns one of the most similar "training documents" it has in its database, verbatim. You could say that this search engine has a "creativity problem" - it will never respond with something new. An LLM is 100% dreaming and has the hallucination problem. A search engine is 0% dreaming and has the creativity problem. All that said, I realize that what people *actually* mean is they don't want an LLM Assistant (a product like ChatGPT etc.) to hallucinate. An LLM Assistant is a lot more complex system than just the LLM itself, even if one is at the heart of it. There are many ways to mitigate hallcuinations in these systems - using Retrieval Augmented Generation (RAG) to more strongly anchor the dreams in real data through in-context learning is maybe the most common one. Disagreements between multiple samples, reflection, verification chains. Decoding uncertainty from activations. Tool use. All an active and very interesting areas of research. TLDR I know I'm being super pedantic but the LLM has no "hallucination problem". Hallucination is not a bug, it is LLM's greatest feature. The LLM Assistant has a hallucination problem, and we should fix it. </rant> Okay I feel much better now :)