Tootfinder

Imagine ChatGPT but instead of predicting text it just linked you to the to 3 documents most-influential on the probabilities that would have been used to predict that text.
Could even generate some info about which parts of each would have been combined how.
There would still be issues with how training data is sourced and filtered, but these could be solved by crawling normally respecting robots.txt and by paying filterers a fair wage with a more relaxed work schedule and mental health support.
The energy issues are mainly about wild future investment and wasteful query spam, not optimized present-day per-query usage.
Is this "just search?"
Yes, but it would have some advantages for a lot of use cases, mainly in synthesizing results across multiple documents and in leveraging a language model more fully to find relevant stuff.
When we talk about the harms of current corporate LLMs, the opportunity cost of NOT building things like this is part of that.
The equivalent for art would have been so amazing too! "Here are some artists that can do what you want, with examples pulled from their portfolios."
It would be a really cool coding assistant that I'd actually encourage my students to use (with some guidelines).
#AI #GenAI #LLMs

LAPIS: A Performance Portable, High Productivity Compiler Framework
Brian Kelley, Sivasankaran Rajamanickam
https://arxiv.org/abs/2509.25605 https://arxiv.…

LAPIS: A Performance Portable, High Productivity Compiler Framework
Portability, performance, and productivity are three critical dimensions for evaluating a programming model or compiler infrastructure. Several modern programming models for computational science focus on performance and portability. On the other end, several machine learning focused programming models focus on portability and productivity. A clear solution that is strong in all three dimensions has yet to emerge. A second related problem arises when use cases from computational science converg…

"Big Oil isn’t part of the clean energy push, despite its claims, study shows"
#FossilFuels #Climate #ClimateChange

Big Oil isn’t part of the clean energy push, despite its claims, study shows
A new study that mapped the portfolios of the world’s 250 biggest oil and gas companies found their deployment of renewable energy is paltry: they’re responsible for just 1.42% of the global renewable energy capacity in operation. Despite announcing ambitious plans to embrace renewables, a mere 0.1% of the primary energy they produce comes from […]

Big Oil isn’t part of the clean energy push, despite its claims, study shows https://news.mongabay.com/short-article/2025/10/big-oil-isnt-part-of-the-clean-energy-push-despite-its-claims-study-shows/

Big Oil isn’t part of the clean energy push, despite its claims, study shows
A new study that mapped the portfolios of the world’s 250 biggest oil and gas companies found their deployment of renewable energy is paltry: they’re responsible for just 1.42% of the global renewable energy capacity in operation. Despite announcing ambitious plans to embrace renewables, a mere 0.1% of the primary energy they produce comes from […]

Happy International #Coffee Day!

Enhanced Pebble Drift Across Planet-Opened Gaps in Windy Protoplanetary Disks
Lorraine Nicholson, Jaehan Bae
https://arxiv.org/abs/2510.07428 https://arxiv…

Enhanced Pebble Drift Across Planet-Opened Gaps in Windy Protoplanetary Disks
When a giant planet forms in a protoplanetary disks, it carves a gap around its orbit separating the disk into two parts: inner disk and outer disk. Traditional disk accretion models, which assume material transport is driven by viscosity, reveal that the planet-induced gap acts like a filter which blocks large dust grains from flowing into the inner disk. However, there is growing evidence that material transport may be driven by magnetically-driven winds instead. By carrying out a suite of tw…

* Step through rendering code and see individual parts draw to the screen like RenderDoc or Unity's Frame Debugger.
* Run to breakpoint in reverse (or data breakpoint!!!)
* Stores a session file that tracks all of your input and code changes that's portable so you can send to coworkers.

Joint Neural SDF Reconstruction and Semantic Segmentation for CAD Models
Shen Fan, Przemyslaw Musialski
https://arxiv.org/abs/2510.03837 https://arxiv.org/…

Joint Neural SDF Reconstruction and Semantic Segmentation for CAD Models
We propose a simple, data-efficient pipeline that augments an implicit reconstruction network based on neural SDF-based CAD parts with a part-segmentation head trained under PartField-generated supervision. Unlike methods tied to fixed taxonomies, our model accepts meshes with any number of parts and produces coherent, geometry-aligned labels in a single pass. We evaluate on randomly sampled CAD meshes from the ABC dataset with intentionally varied part cardinalities, including over-segmented s…

Infall Explains the Disk Kinematics of AB Aur Without Gravitational Instability
Josh Calcino, Daniel J. Price, Chris W. Ormel
https://arxiv.org/abs/2510.05601 https://

Infall Explains the Disk Kinematics of AB Aur Without Gravitational Instability
Late-stage infall onto protoplanetary disks can produce large scale spiral arms. In this paper we used 3D smoothed particle hydrodynamics and radiative transfer simulations to study the kinematic perturbations induced in disks by infalling material. We found that deviations from Keplerian rotation are predominantly in the radial and vertical velocity components, spatially correlated with spiral arms in the gas surface density. The infall produces observable wiggles in the channel maps, analogou…