I don't have any experience in this, but would be curious to see what happens if you make a couple of screenshots of your current website, pack you source code into a txt file and paste it into gemini pro https://aistudio.google.com asking to get some feedback on how to improve, and then ask to implement it (maybe using claude [code?], it may work better)
Yes of course! DataBridge is a multi-modal database with the belief that retrieval over unstructured data - such as videos, pdfs, and other documents should work with the same reliability, speed, and consistency as regular database data structures (and at a similar level of abstraction). Right now, we're focusing on helping simplify RAG pipelines for users.
multi-modal database means that we treat multi-modal data, such as images, audio, and video as first-class citizens and provide natural language search over them. That is, you can ingest multi-modal content into DataBridge the same way you'd ingest structured data into a database. You can perform updates over this information, extract metadata, or define custom parsing/processing rules (eg. redact any PII).
Your search queries would go through a planner which - depending on the kind of data we're retrieving - will call the correct tools to extract information from the data and respond to your query.
For instance, this could be function calling over object-tracking data if your query relates to object movements over a video. This could also be a call to ColQwen in case we're looking for particular features within a diagram-heavy PDF. It could also be a simple semantic search if thats what the planner deems most useful.
The idea is that traditional databases work the same way - query planning systems figure out the best path to execute the user query, and pass that to the query execution engine. We think a lot of this complexity can be abstracted away from the user - as long as we can provide them strong retrieval guarantees (the same ways Databases have SLAs).
Oh I'm trying to build something similar, let's chat! I've started with identifying potential project from ICLR2025, but it has an "entrepreneur" part in the response https://openreview-copilot.eamag.me/
Matter refers to particles or collection of particles that have mass+volume. These particles can be arranged or behave in different ways, and that is roughly what a "state of matter" is. You know how in solid all the atoms are fixed, but in a gas atoms/molecules are flying about.
There are in fact other forms of matter. In plasma you just have ions (instead of atoms/molecules) just zipping about. In neutron stars, you have pretty much only neutrons collapsed into a packed ball.
You can also make systems at higher levels of abstraction that have some of this matter or particle like behavior. A simple example is "phonons", which are a small packet of vibration (of atoms) that travels inside a solid much like a photon travels through space. I think phonons don't have a "mass", so they are not matter.
Here, they construct a quantum system, some of whose degrees of freedom behave like a matter particle. Qubits are then made from the states of this particle.
For example in some cases they have a "gas" of electrons. It's not a normal gas that you can put in a balloon, it only can live inside a solid. If you ignore the atoms in the solid, in some cases the electrons are free enough to think they are a gas. That is similar enough to a normal gas, and then they just call it a gas.
Sometimes the interesting part is a surface between two semiconductors, so they may have a 2D gas. (I'm not sure if this experiment is in 2D or 3D.)
Sometimes the electrons make weird patterns, that are very stable and move around without deformation, and they will call it a quasiparticle, and ignore that it's formed by electrons, and directly think that it's a single entity. And analyze how this quasiparticles apear an disappear and colide with other particles. It's like working on a high level of abstraction, to make the calculations easier. [2]
In particular, if you arrange the electrons very smartly, they create a quasiparticle that is it's own anti-quasipartilce. In particular, this is a Majorana quasiparticle.
This is somewhat related to topological properties of the distribution of the properties of the electrons. Were topological means that is stable under smooth deformations and that helps to make it also stable under thermal noise and other ugly interference. But this is going too far from my area, so my handwaving is not very reliable.
[1] They probably think my area is weird, so we are even :) .
[2] Sometimes the high level abstraction is not an approximation.
No idea. I read all of that during the Physics courses in the university. The examples are dispersed in a few courses.
I found this video by minutephysics with more details https://www.youtube.com/watch?v=KbsnY--LFh0 It's not super technical, but I think it my clarify my explanation and add more details.
And, as always(?), 3Blue1Brown also contributes https://www.youtube.com/watch?v=IQqtsm-bBRU"This open problem taught me what topology is" I'll warn that while it is illustrative, and only 28 minutes, it's some heady stuff
I think you're just not the target audience. If AI can come up with some good ideas and then split it into tasks some of them an undergrad can do - it can speed up the global research speed by involving more people in useful science
In science, having ideas is not the limiting factor. They're just automating the wrong thing. I want to have ideas and ask the machine to test for me, not the other way around.
If I understand what's been published about this, it isn't just ideation, but also critiquing and ranking them, to select the few most worth pursuing.
Choosing a hypothesis to test is actually a hard problem, and one that a lot of humans do poorly, with significant impact on their subsequent career. From what I have seen as an outsider to academia, many of the people who choose good hypotheses for their dissertation describe it as having been lucky.
I bet all of these researchers involved had a long list of candidates they'd like to test and have a very good idea what the lowest hanging fruit are, sometimes for more interesting reasons than 'it was used successfully as an inhibitor for X and hasn't been tried yet in this context' — not that that isn't a perfectly good reason. I don't think ideas are the limiting factor. The reason attention was paid to this particular candidate is because google put money down.
The difference is the complexity of ideas. There are straightforward ideas anyone can test and improve, and there are ideas where only PhDs in CERN can test
I don't think that's really right. E.g. what makes finding the Higgs boson difficult is that you need to build a really large collider, not coming up with the idea, which could be done 50 years earlier. Admittedly the Higgs boson is still a "complex idea", but the bottleneck still was the actual testing.
reply