Skip Navigation

InitialsDiceBearhttps://github.com/dicebear/dicebearhttps://creativecommons.org/publicdomain/zero/1.0/„Initials” (https://github.com/dicebear/dicebear) by „DiceBear”, licensed under „CC0 1.0” (https://creativecommons.org/publicdomain/zero/1.0/)B

bss03

@ bss03 @infosec.pub

Posts
2
Comments
614
Joined
3 yr. ago

  • POV: M$ releases a new VSCode version

    Jump

  • it was inherent to the terminal that you can’t position the cursor and select text using the mouse, and also inherent that there are not right-click menus

    There may be terminals for which that is a limitation, but with neovim in konsole, I can (left) click to move the cursor or select text and right-click to get a contextual menu.

    Separately, the TUI libary we use for RESTMan is Brick and does support mouse events. IIRC we only support mouse-wheel scrolling, but there's a lot more that is possible.

    So, I think it might worth your time to try some TUI alternatives to your GUI environment.

    Your larger point that TUI can't fully replace GUI stands, at least IMO. Even for plain text formats (e.g. dotty) that I want to edit with a fixed-width font, an image (pre)view can be essential to some tasks. I don't usually find that's true for coding tasks, but design docs often benefit from charts.

  • POV: M$ releases a new VSCode version

    Jump

  • You might in the future. Both vim and neovim are accepting genAI code contributions. There's a vim-classic fork around, but it is "rolled back" to pre-vimscript9.

  • Claude Code's source code appears to have leaked: here's what we know

    Jump

  • Once companies started suing people trying to practice "responsible disclosure", I stopped attacking people that choose maximum disclosure.

    Responsible disclosure has always been a bit of a hedge. It's rare to be able to show you are actually the first person/organization to discover a vulnerability.

  • Claude Code's source code appears to have leaked: here's what we know

    Jump

  • we are going to need to develop a different model of learning, using, and processing information that considers the provenance of where the information came from and how it got there

    They used to teach this in schools under "critical thinking skills". Following the chain of sources to the primary sources was a task I had to to (at least in part) more than once in secondary school.

    Authoritarians don't like that tho.

  • Your Engineers Aren't Lazy, Your Codebase Is Punishing Them

    Jump

  • Yeah, but the LLM will eventually realize there's no fixing it and delete everything. /s

    Under capitalism there's no lowest quality level at which workers can refuse to exchange their labor for pay.

  • People who have been on Lemmy since the very initial reddit API debacle, or before even that, how much do you think it has changed?

    Jump

  • More people around bitching that they can't farm karma with their tired opinions.

  • Deleted

    Permanently Deleted

    Jump

  • I actually think you have to go so slow that your (probable) position extends to the other side of the wall. Unfortunately speeds that slow are incompatible with (the processes of) life.

  • Deleted

    Permanently Deleted

    Jump
  • Deleted

    Permanently Deleted

    Jump

  • BTW, before a detector aparatus can be created, many physics results were (are?) identified through observation, which might include a measurement or might be qualitative.

  • Deleted

    Permanently Deleted

    Jump

  • Announcing ARC-AGI-3 - A benchmark that tests if AI can explore, learn, and adapt in unfamiliar situations. Humans score 100%. Frontier AI scores 0.26%.

    Jump

  • Yeah, there was some phonics in my primary school education, and I continue to approach new words in that way sometimes. But, they said Phonetically.

  • Announcing ARC-AGI-3 - A benchmark that tests if AI can explore, learn, and adapt in unfamiliar situations. Humans score 100%. Frontier AI scores 0.26%.

    Jump

  • Cave 1.0 scored 1000000% but also force fed the proxy lemons, so it was treated as a failure.

  • Announcing ARC-AGI-3 - A benchmark that tests if AI can explore, learn, and adapt in unfamiliar situations. Humans score 100%. Frontier AI scores 0.26%.

    Jump

  • Wait, I thought phonetically (example: papa hotel oscar novermber echo tango india charle alfa lima lima yankee) meant using a phonetic alphabet, not using word(s) with the same Soundex encoding.

  • Announcing ARC-AGI-3 - A benchmark that tests if AI can explore, learn, and adapt in unfamiliar situations. Humans score 100%. Frontier AI scores 0.26%.

    Jump

  • If you look at the list of tasks,, you can see how the 4 frontier models did. Some of them did complete one or two levels of one or two tasks. None of them completed a whole task. Some of the reasoning logs are funny in the replays.

  • Announcing ARC-AGI-3 - A benchmark that tests if AI can explore, learn, and adapt in unfamiliar situations. Humans score 100%. Frontier AI scores 0.26%.

    Jump

  • Announcing ARC-AGI-3 - A benchmark that tests if AI can explore, learn, and adapt in unfamiliar situations. Humans score 100%. Frontier AI scores 0.26%.

    Jump

  • Yeah, for a fixed ruleset that can be provided up front the Alpha-Zero approach seems to work great.

    These tasks strike me as a bit different. I'm sure the ruleset is fixed somewhere, but it's not disclosed to the participants. In the task I walked myself through, there was a new wrinkle in each part -- a new interactable, a (more) hidden goal, or an information limit. And, of course, part of the task is "discovering" all that from the bitmap frame(s) provided.

    I'm unconvinced of the hype around "AI", but this does seem like a legitimate research target that might stymie the Alpha{Go,Zero,Fold} series at least a bit.

  • Announcing ARC-AGI-3 - A benchmark that tests if AI can explore, learn, and adapt in unfamiliar situations. Humans score 100%. Frontier AI scores 0.26%.

    Jump

  • The founder of ARC worked at Google until 2024 and wrote 2.5+ books in Deep Learning. So, I expect some of these benchmarks are based on limitations seen in Deepmind.

    That said, it would be interesting to see how well Deepmind does at these tasks. My understanding is that the private tasks would still be dynamic enough to require "on the job training" so an Alpha-Go / Alpha-Zero / Alpha-Fold approach is unlikely to do well on ARC-AGI-3.

    Still, I think commentary around models (including, but not limited to something from Deepmind) attempting these tasks would be much more interesting than most of the discourse around generative AI, whether text, image, video, or code generation.

  • Announcing ARC-AGI-3 - A benchmark that tests if AI can explore, learn, and adapt in unfamiliar situations. Humans score 100%. Frontier AI scores 0.26%.

    Jump

  • Announcing ARC-AGI-3 - A benchmark that tests if AI can explore, learn, and adapt in unfamiliar situations. Humans score 100%. Frontier AI scores 0.26%.

    Jump

  • I couldn't find replays. Are there more? Also, it is a bit funny that "building the bridge" which at one point seems to be Claude's "chosen goal" is just "running out of moves" and failing the task.

    Task failed successfully, Claude. Task failed, successfully.