This line stuck out to me as well, but my follow up thought was different.
I’ve had friends who have been on cocktails like these, and one of them once said something like, “I’ve been depressed before, and this is not that. I’m not depressed. I don’t have the emotional capacity to be depressed. This is more like a total emotional blank slate.”
She was basically a robot for a few months. Incapable of really any emotions, including sadness, anxiety, frustration, etc. Suffice to say, she also didn’t have the emotional drive to push her towards positive things like deciding on how to spend her weekend free time.
Thankfully she’s changed her meds and is feeling overall better (if, admittedly, at the price of some emotional stability).
One of my favorite applications of multimodal LLMs thus far is the ability to:
1. Draw a DAG of whatever pipeline I’m working on with pen and paper.
2. Take a photo of the graph, mistakes and all.
3. Ask ChatGPT to translate the image into mermaid.js
Given how complicated the pipelines are that I’m working with and the sloppiness of the hand drawn image, it’s truly amazing how well this workflow works.
Yep, I'm in the rare disease space. "impossible" is pretty appropriate.
It's tricky. On the one hand, it's obviously not appropriate to be flippant about patient privacy. On the other, it's clearly that advancements in human health are being hindered by our current approach to (dis)allowing researchers access to data.
I want to second this. It seems like document chunking is the most difficult part of the pipeline at this point.
You gave the example of unstructured PDF, but there are challenges with structured docs as well. We’ve run into docs that are hard to chunk because of this deeply nested and repeated structure. For example, there might be a long experimental protocol with multiple steps; at the end of each step, there’s a table “Debugging” for troubleshooting anything that might have gone wrong in that step. The debugging table is a natural chunk, except that once chunked there are a dozen such tables that are semantically similar when decoupled from their original context and position in the tree structure of the document.
This is one example, but there are many other cases where key context for a chunk is nearby in a structured sense, but far away in the flattened document, and therefore completely lost when chunking.
Just to add to the list of this Jim Simons did and funded, he also established the Simons Foundation Autism Research Initiative (SFARI).
"SFARI’s mission is to improve the understanding, diagnosis and treatment of autism spectrum disorders by funding innovative research of the highest quality and relevance."
SFARI in turn funds a lot of foundational neurological and rare disease research, since autism is such a common phenotype.
The paper kinda leaves you hanging on the "alternatives" front, even though they have a section dedicated to it.
In addition to the _quality_ of any proposed alternative(s), computational speed also has to be a consideration. I've run into multiple situations where you want to measure similarities on the order of millions/billions of times. Especially for realtime applications (like RAG?) speed may even out weight quality.
I used to run into this problem all the time in grad school. Once a month or so I'd load a data set, do some dumb Python operation on it that took significantly more memory than I predicted, and BAM! I'd have to restart my laptop.
I just kinda assumed that's how computers worked until I got a Mac a couple of months ago...
The link suggests that there might be some default parameters you could change to protect against this behavior. Does anyone have any suggestions on what settings to change?