They are not really token-in token-out per se, they are embedding-in embedding-out.
When operating on text, you embed each token into the LLMs embedding space. You go from a discrete token to a point in embedding space.
Likewise, when processing images, you have a image embedding model which produces a set of embedding vectors representing the contents of the image in the LLMs embedding (latent) space.
This same concept can be extended to compaction. Instead if limiting yourself to discrete tokens, you could generate a set of embedding vectors which represent the contents of the compacted conversation in latent space.
These have the possibility of containing a lot more semantic information per vector, which is why this can be appealing.
A big downside is decreased interpretability. AI safety people are generally fairly opposed to latent space reasoning for example, it can be harder to tell what the model is actually doing and if it is trying to deceive you.
It seems likely to me this was driven by the `ultra` mode in 5.6, which fans subagents to do work. This mode was previously only available in the web UI (what was previously known as pro?)
It seems possible they trained this by doing full RL rollouts of agents interacting with each other. They likely view these prompts somewhat the same as raw reasoning traces, they don't want people to train directly on them.
I am unsure if this has been confirmed, but there are some signs that the opaque "compaction blob" they return from their dedicated compaction endpoint might not be text at all, rather a latent space representation of the conversation. The fact that OpenAIs compaction seems to be much higher fidelity than a lot of other providers makes me inclined to believe this.
If this is true, it doesn't seem far fetched to infer that they might be applying similar techniques to prompting subagents.
I would be curious to see if this way of spawning subagents (encrypted blob) is used when subagents of a different model type is spawned.
What makes it so great for me is the effortlessness.
I often use Python for quick one off scripts. With UV I can just do `uv init`, `uv add` to add dependencies, and `uv run` whatever script I am working on. I am up and running in under a minute. I also feel confident that the setup isn't going to randomly break in a few weeks.
With most other solutions I have tried in the Python ecosystem, it always seemed significantly more brittle. It felt more like a collection of hacks than anything else.
Your dismissal of this research shows a fundamental misunderstanding of how scientific analysis works.
The study analyzed two separate phenomena - changes in engagement for Musk's posts AND changes in engagement for Republican content - occurring simultaneously. When taken together, these patterns strongly suggest algorithmic changes, not just organic user behavior from the assassination attempt.
Saying "you can't do that for the reasons I mentioned" is just wrong. Studies absolutely can point toward likely explanations without 100% certainty - that's literally how science progresses. The researchers used appropriate cautious language because they understand scientific rigor, not because their analysis is "crap."
Your argument that "if circumstances don't allow for a proper study, you don't make it" would eliminate most scientific advancement. Should we have abandoned Alzheimer's research because perfect data wasn't available? Obviously not.
What's truly absurd here is your selective skepticism. You demand impossibly high standards of proof for findings you dislike while accepting "common sense" explanations that align with your preconceptions.
Before dismissing research as "garbage" that "causes division," maybe consider whether your reaction is based on methodological concerns or simply that the evidence contradicts your preferred narrative about Musk. Your eagerness to defend him while offering nothing but personal opinion suggests it's the latter.
It seems to me that when taken together, these two make a fair case that there was a change in algorithm on the given date?
Besides, the study you are claiming "is not science" did not even make strong claims as to there being an algorithm change. The study made an analysis, and concluded that they might point towards there being an algorithmic change.
Additionally, as you say "If the claim is that Musk tuned the algorithms at that special date, you have to prove it by other means". What other means exactly? We have no data to go on except observations of what happens on the platform. From the information we have it seems like everything points towards this being the case.
The case for manipulation happening furthered even more by Elon Musk repeatedly showing he is full of shit, and has no qualms lying and totally making stuff up. (see his repeated lies about Autopilot, his lie about being world class at video games, his lies about DOGE cuts, etc).
Before dismissing these findings, ask yourself honestly: would you apply the same rigorous standards of proof if the algorithmic changes benefited different political figures or viewpoints?
We were talking about his son, he let the process play out in full which is beyond respectable.
When the incoming administration has shown extraordinary will to persecute political opponents, I think it would be unethical not to preemptively pardon these people.
I don't know about this at any detailed level, but doesn't designing standard cells for leading edge nodes involve a lot of trial and error? Is a lot of the issues that can occur even well understood to the level that it can be simulated?
With the approach you mention, would it involve creating "custom standard cells", or would the software allow placement of every transistor outside of even a standard cell grid? If the latter, I would have trouble believing it could be feasible with the order of magnitude of computing power we have available to us today.
It's probably more of a node thing than a fab thing. You would have a much easier time getting the fab to do random stuff for you on a legacy node compared to a leading edge node.
Leading edge nodes are basically black magic and are right on the edge of working vs producing broken chips.
You as a customer would never want to be in a position where you are solely responsible for yields.
For cases like this, you just need to convince it that it would be inappropriate to generate anything that does not follow your instructions. Mention how you are planning to use it as an avatar and it would be inappropriate/cultural appropriation for it to deviate.
> We are writing to inform you that we have discovered two Home Assistant integration plug-ins developed by you [...] that are in violation of our terms of service
> Specifically, the plug-ins are using our services in an unauthorized manner, which is causing significant economic harm to our Company.
> We take the protection of our intellectual property very seriously and demand that you immediately cease and desist all illegal activities related to the development and distribution of these plug-ins.
They seem to be threatening legal action because he is violating their terms of service? This doesn't make much sense to me
[ my public key: https://keybase.io/hansihe; my proof: https://keybase.io/hansihe/sigs/PfpwSyBwKYgn8cDVS0FghzCw2sZF1N_ozR08IekKMPE ]