Tootfinder

Have u created a small being today?

#Oumuamua was first spotted on October 19, 2017, by a telescope in Hawaii called Pan-STARRS 1.
At first, scientists thought it was just a regular comet or asteroid -- But as they watched it more closely, they realized it was very different.
Here are some key things about Oumuamua:
#Shape: Unlike most as…

What is Oumuamua and Why Did it Puzzle Scientists? - Astronex
Imagine looking up at the night sky and seeing something zoom by, something unlike anything you've ever seen before. It's not a star, it's not a planet, and

Series A, Episode 12 - Deliverance
ENSOR: I'm on three-quarter boost as it is. She's not responding! I'm going to maximum. It's all right. It's all right. She's slowing. Compensators beginning to hold. [Ship starts to steady.] Come on, come on, that's my beauty. That's...come on. Pull us back, pull us back. All right, she's coming back. We're all right.
MARYATT: Don't do that too often, will you? I'm a very nervous passenger.

Ah yeah! @…’s horror advent calendar has arrived 🙌
Here’s a little taste from last year..

Or just use a solid rectangle selection tool and DELETE whatever you want to redact.
I’ve never understood how people can be so terrible at redacting pixel-based images. PDFs can be a special case (hint: redact, then Print as PDF)
https://todon.eu/@cf/115521875990737714

Crimson Fork /🌌🏳️‍⚧️&ΘΔ (@cf@todon.eu)
Attached: 1 image When censoring names or other things in screenshots/images, it's important to know that a lot of image editing programs default to a brush with partial transparency both for the shape of the brush as well as sometimes the color itself, We've personally been able to extract enough information from censored screenshots to narrow down what the crossed out sections are, even if to the naked eye they were not readable. Built in mobile image editors tend to be the worst offender in…

You Only Train Once: Differentiable Subset Selection for Omics Data
Daphn\'e Chopard, Jorge da Silva Gon\c{c}alves, Irene Cannistraci, Thomas M. Sutter, Julia E. Vogt
https://arxiv.org/abs/2512.17678 https://arxiv.org/pdf/2512.17678 https://arxiv.org/html/2512.17678
arXiv:2512.17678v1 Announce Type: new
Abstract: Selecting compact and informative gene subsets from single-cell transcriptomic data is essential for biomarker discovery, improving interpretability, and cost-effective profiling. However, most existing feature selection approaches either operate as multi-stage pipelines or rely on post hoc feature attribution, making selection and prediction weakly coupled. In this work, we present YOTO (you only train once), an end-to-end framework that jointly identifies discrete gene subsets and performs prediction within a single differentiable architecture. In our model, the prediction task directly guides which genes are selected, while the learned subsets, in turn, shape the predictive representation. This closed feedback loop enables the model to iteratively refine both what it selects and how it predicts during training. Unlike existing approaches, YOTO enforces sparsity so that only the selected genes contribute to inference, eliminating the need to train additional downstream classifiers. Through a multi-task learning design, the model learns shared representations across related objectives, allowing partially labeled datasets to inform one another, and discovering gene subsets that generalize across tasks without additional training steps. We evaluate YOTO on two representative single-cell RNA-seq datasets, showing that it consistently outperforms state-of-the-art baselines. These results demonstrate that sparse, end-to-end, multi-task gene subset selection improves predictive performance and yields compact and meaningful gene subsets, advancing biomarker discovery and single-cell analysis.
toXiv_bot_toot