what I'm finding is that, the models want to be more of an artist than a replacement for photoshop — which is fine, but I want to be the artist here, and want the tool to be more of a "magically easier photoshop where I ask it what to do in detail, and it does that — not more
@soumithchintala
-
Image Generation Models Struggle With Precise Localized Edits
By
–
i'll give a representative but not exact example: change the color of X's shirt from blue to red: the generations often change the entire shirt style itself — they don't respect how much and what I'm trying to change, and dont try to preserve details I ask to preserve
-
AI Image Transformation Tools: Selective Detail Preservation
By
–
what are AI products that allow me to transform existing images, while preserving some selective details (that i select), like faces, areas, etc.? the tools I've used so far only take the selection as a hint, or dont generate well around the selection? trying personalized art
-
Cost-effective AI model training under $6 million
By
–
Looks pretty good, considering the cost-to-train is under $6m. Needs a vibe-check.
-
How do xAI employees balance intense work schedules with family life?
By
–
i'm curious (with all good intentions), how do @xai employees who have kids manage things, if you're grinding like 15 hour days every day including weekends and holidays?
Elon obviously has like a hundred nannies, but how do the normal employees manage?
the simplest answer is -
ARC Benchmark Limitations: Smoke Tests vs Real AI Progress
By
–
that's largely been my opinion on beating ARC. Over the past decade, I've seen too many artificial benchmarks becoming the target, and being beaten, and resulted in no or little meaningful actual progress.
I guess it can be a smoke test, but I don't think it's going to be a -
User adopts Perplexity for AI-powered research with citations
By
–
i usually ask perplexity these days and it does a good job mostly, with references.
-
Nvidia Shield delivers top TV streaming performance
By
–
nvidia shield connected to the TV. top performance.
-
o3 Model Architecture: Matrix Multiplications and Gradient Descent Training
By
–
o3 is a lot of matmuls trained with gradient descent.