Topfi

7 ms·

91.
by Topfi 2mo ago
Yeah, agree, purely from a user expectation perspective, strict sounds like it'd suppress everything, not just third-party, especially with lower compatibility turned on. Yeah "browser.events.data.microsoft.com" is in, but ma
92.
by Topfi 2mo ago
Just saw the edit, all clicked now. As described this is expected behavior. The entitieslist (can't link right now because Github is down but got a local copy for some experiments) contains exceptions for owners of tracking URLs, in th
93.
by Topfi 2mo ago
Github being under the CoreAI division probably also doesn't help the engineers prioritize addressing infrastructure issues and makes using LLM load as an excuse feel self-inflicted. Akin to feeling sorry when a pyromaniacs house burns
94.
by Topfi 2mo ago
No one ever has to commit anything time sensitive.
95.
by Topfi 2mo ago
Standard, Strict or Custom? Strict should block that one for sure, will check once at home.
96.
by Topfi 2mo ago
Fair point, for text it is far harder to prevent signatures being applied to generated text vs images at the moment of capture and most approaches I can come up with to remedy this can either be bypassed (edit histories can be output by mod
97.
by Topfi 2mo ago
Why not switch it around? Solves problems of privacy (no data upload), is far more doable and it's far more important to be able to verify content has not been tampered with after originating from a human or capture device (like a came
98.
by Topfi 2mo ago
I had sent this an hour ago as a quote with a few others (lactose bigotry, of course the kidney disappointment) to someone via iMessage. The Apple Intelligence summary was unfortunately not screenshotted by them, but I am told it was a dooz
99.
by Topfi 2mo ago
> This is false. Perhaps you meant they have insufficient methods, or imperfect methods, but asserting none at all is facile and your links do not say that at all. I did say "no reliable internalised way to assess the accuracy of th
100.
by Topfi 2mo ago
Greatly varies by the model, but such a flat denial along with calling those with (verifiably accurate [0]) experiences which happen to be different from yours "the anti-AI crew" and just assuming they can't possible use LLMs
101.
by Topfi 2mo ago
> [...] not make it a bit more explicitly known. Honest question, how could it be made more explicitly known? It's literally under ACKNOWLEDGMENTS, in the most clear language possible.
102.
by Topfi 2mo ago
Deepseek v4 Pro or GLM 5.3 for software architecture I feel are only deployed for general, rather doubtful those are for narrow stuff, if we are sticking with this threads mentioned models. For small models, sure, narrowly targeted sets whi
103.
by Topfi 2mo ago
Anything from ML pipelines for language specific pruning over a Rust/JS/CSS mix codebase to assistance in motorcycle maintenance and different canvas coatings. Most of my evals build on those requirements and especially past failu
104.
by Topfi 2mo ago
Honest question, how do you assess models this quickly? What metrics are you using? Would love to get my suite from multiple days and hundreds of prompts down to minutes. Got a few first pass tasks I run upon release for an initial experien
105.
by Topfi 2mo ago
Sure, but you then only are on Google Maps, Search and other services by them. Have your own website and be indexed by everyone from Kagi, over Bing to DDG.
106.
by Topfi 2mo ago
That quite strong "no one" surprised me so I checked and looking through a few blog posts from back then, they did advertise gpt-3.5-turbo as a straight up improvement and, once text-davinci-003 was to be deprecated, the instruct
107.
by Topfi 2mo ago
Still, what'd be the advantage? There is a lot, a massive delta, a chasm still to improve before there is even a theoretical potential re. performance. GPT-5.6-Sol still writes JS in inefficient and roundabout ways that leave a lot on
108.
by Topfi 2mo ago
No quality compromise/degradation is something I have had this industry, including especially OpenAI, claim multiple times in the past and I have more than once been able to verify that it was in fact not the case. Examples being gpt-3
109.
by Topfi 2mo ago
A great future for us all to look forward to. Human readability gone, any changes must go through a providers API so cost+restrictions are a constant and we'll need far more tokens for even simple operations. Call me old fashioned, sti
110.
by Topfi 2mo ago
Unless I have read over it, besides the animation in the intelligence vs speed graph which only mentions internal data and not whether they truly reran the AA suite, there is no actually solid statement on the important aspect of performanc
111.
by Topfi 2mo ago
> What's new in Gemini 3.7 Flash [0] > Coding and agentic tasks: Significantly higher quality on real-world software engineering and agentic benchmarks, improving issue resolution and reducing failed agent loops. > Web develop
112.
by Topfi 2mo ago
Where do you think Google and co. source that data from? Having an actual, searchable, up to date menu is, just for a11y and accuracy far more valuable than someone’s poorly lit, low res upload of an photo taken from the menu circa 2019, ju
113.
by Topfi 2mo ago
Looking forward to the release, for testing and learning how their design differs from other labs alone. Similar to Inkling, wouldn't be surprised if their first public efforts will fall short of hype but have amazing potential and som
114.
Load What You Need: Smaller Versions of Mutililingual Bert (2020) [pdf](aclanthology.org)
1 points by Topfi 2mo ago|0 comments
115.
Chatgpt.com – Degraded Performance(status.openai.com)
1 points by Topfi 2mo ago|0 comments
116.
by Topfi 2mo ago
Couldn't agree more and can only recommend T5 as a base to anyone. It's amazing to get started, whether as a learning resource or for real (albeit very tailored) applications. Especially the BigScience fine tunes are such a great
117.
by Topfi 2mo ago
That goalpost has moved a few miles in two comments.
118.
by Topfi 2mo ago
Developers are just a subset of users. If Apple e.g. didn't allow third-party apps to access the camera for more granular control over photography, we wouldn't differentiate between user hostile and photographer hostile either, be
119.
by Topfi 2mo ago
Sure you can, via e.g. Termux. Have done so. Also, I don't want to run Xcode, it is arguably the worst development environment and hasn't gotten any better over the years. I have to use Xcode regardless, if I want to develop for i
120.
by Topfi 2mo ago
Still pending, or am I mistaken? Can anyone with knowledge on the US process make an estimate how likely this is to go through? I know of Uniloc and their "efforts", is there a possibility that this could take a similarly damaging

More ›