avereveard
avereveard on Hacker News
Is this a pro 3.1 situation that by the time ot reached the users it produced no more work than the flash variant or competitor for absurd pricing?
Because they don't get held responsible enough for bypissing procedures. Some system are too complex for fully automated operarions, and the human need to play along.
Safety culture is blameless, but only if everyone plays along.
Dispatcher, conductor at that time could not have know that dispatcher ordered manual bypass ignorig the rest of the guardrails of a traffic resume procedure
It does because the claims that AI killed it are wrong. Would have died from AI anyway probably yes but it doesn't make the AI scapegoat any more true. Why do you have such an interest in blaming AI that you argue from a false position just because it might have happened?
Why do worker accept to work in such conditions? Its like driving tired every day. Eventually you fuck up. Where was the union?
List european countries with summer provision for friday half off
AI promised less workers, I guess.
"oh no, not the creative commons resources we distribute freely even if they contain copyrighted material!"
Again this is the second time this year
Yeah a lot of people worry of ai in high stake environment but a jev-like evaluating a resume or a university application is much more impactful and likely to happen
TAA introduce aliasing on fast moving element which is far more annoying than blurring
No because it doesnt get backgrounded and I would need to hold one end of the stream at all time, and feature completedness is spotty
I use the protocol claude code terminal use to talk to its server to spawn agents, opencode server protocol which is published for the same purpose, and manage pi command lines by pid
Claude code agents have a protocol that's usable for the same purpose it just listen locally but it's usable i have my own claw-like that has options for claude, codex, and pi (my subscription, wife subscription, and i got a yearly sub on glm when it was cheap)
> what exactly doesn’t work in the old product
Assumptions. I've rebuilt a number of inference engines to run on windows and unify the dependency venv to last version to save space, unrolling model blocks so they fit in my 16gb card etc. A lot of thing thay I was locked out are possible now as i am no longer held back by invariants decided by someone else, and because I do not have to spend time to get things done. I've changed games to have a less grindy loop, added features, or making learning them easier. Ive fixed drivers, patched versions, brought dlss5 to game weren't supposed to have it
It's not ground breaking, but its liberating.
Can you find vc appetite for a startup before they buy equity
Not even a dlc and cosmetics?
About half a second per decision on six cores
I've stumbled upon a reliablish pipeline you create a reference sheet and a single image pose then trellis v2 for the body and unirig for tigging, then you generate the poses you need in a sheet and send the pose sheet plus the reference images, and use these as 2d anymation cycles and you compose on top of the scene with lanes, this focus all attention of the model to fidelity instead of background integration
It's not about the challenge tbh i play rail worlds because I like trains and while it evens out long term the start is slower and I dont feel like having time based evolution is a fair or interesting. I do leave pollution based evolution tho. That feed into base design choices
> new name
Swagger was the new old name for the concept
Theres transport and context management. A cross agent status cache with semaphore over a testing harness driving a browser that can rewind and retry is much easier for agents to drive from mcp than selenium or whatever is fashionable these days. Didn't replace unit testing per se but to rca and fix it's a much better tool.
I disable expa sion but also biters evolution. Without expansion and evolution biters are just the right amount of pressure for me for casual play without completely losing having to climb half of the game tech tree
This is such a 2024 take. Model will push back and tell you the knowledge gap they have, but people are just used to ask gooogle leading questions, which port badly to llm as it makes them defend the position instead of research data
what worked for me is a frontend driving the agent, capturing every user message, every commit, every pull request, putting them in a graph (uses a frontend because I didn't want to load a coding agent with tools + responsibility of book keeping) and the agent get tools to search reason behind code changes and see the high quality user input underneath instead of the sloppy self written justifications
Calibrated probability across multi task with zero shot I guess. A reranker is single task and tuning it make it even more narrow. And I guess some piping to make multiclass efficient since you cannot mask logprob for independent questions in the same output space without throwing calibration away.
entertainment hasn't consumed art in a while most of it is mass produced at minimum cost and zero artistic freedom. some sparingly rare project that could be called art remains, driven by passion of the craft, and these aren't going away
btw that how mmlu score things to answer question instead of producing all the answer token they look at logprob of a b c d keys in 2020 making this technique old as dirt in nlp