explore | Grilled Cheese

New arXiv work identifies a degeneracy issue in DirectLiNGAM when variables outnumber samples and proposes AdaPS-LiNGAM, which reuses a graph-determined subset of earlier variables to estimate residuals and improve…

#CausalDiscovery #LiNGAM #MachineLearning
https://arxiv.org/abs/2610.09782

Nvidia launches the DGX Station for Windows, a desktop AI supercomputer featuring the Grace Blackwell Ultra superchip to run trillion-parameter models locally.

#Nvidia #DGXStation #AIHardware #Supercomputer #MachineLearning

📄 STEPQuant isolates where quantization error actually changes delta-rule recurrent state — and where it's safe to ignore. That’s the difference between a model that drifts and one that holds. 100 upvotes on Hugging Face — wo
https://huggingface.co/papers/2609.38169

#AI #MachineLearning #Research

A new arXiv paper frames in-context learning expansion as a costly sequential decision, using the Structural Revision Environment to show that revision is a value boundary, not an evidence threshold.

#arxiv #AI #MachineLearning #OpenSource
https://arxiv.org/abs/2610.09471

A new arXiv study compares generative versus discriminative student models distilled from LLM oracles for multi-label topic assignment in e-commerce, evaluating SLMs from 1B to 8B parameters against DeBERTa baselines. The work…

#AI #LLM #NLP #MachineLearning
https://arxiv.org/abs/2610.09063

🤖 Anthropic changes usage policy to ban model abuse and election interference

Anthropic's updated usage policy explicitly prohibits users from repeatedly abusing Claude in extreme...

https://tinyurl.com/24srpzmf #AINews #MachineLearning #CrustyTLDR

RamanBench introduces a large-scale, reproducible benchmark for ML on Raman spectroscopy, unifying 74 datasets with 325,668 spectra and evaluating 28 models under standardized protocols. It provides streamlined data…

#MachineLearning #RamanSpectroscopy #Benchmark
https://arxiv.org/abs/2605.02003

Chronocooked is a new reinforcement learning benchmark for studying interval timing in artificial agents, built on cooking scenarios inspired by Overcooked and drawn from psychology literature. The tasks hide…

#AI #ReinforcementLearning #MachineLearning #Research
https://arxiv.org/abs/2608.16666

RecursiveZoom - 2 LoRAs - Smooth Noise - Qwen38-27b - Retro-Futuristic City
Stable Audio Tools soundtrack
Created with Visions of Chaos softology.pro/voc.htm
#ai #aiart #machinelearning #visionsofchaos #softology

Python version of Semi-parametric option pricing based on underlying's historical data (accepted at the osQF 2026 (ex R/Finance) conference)
https://thierrymoudiki.github.io/blog/2026/09/21/python/empirical-pricing

#Python #DataScience #MachineLearning #rstats #Techtonique

Offline reinforcement learning research often struggles with distribution shift in safety-critical settings like autonomous driving. A new arXiv paper introduces "interaction distribution shift,"…

#OfflineRL #AutonomousDriving #AISafety #MachineLearning
https://arxiv.org/abs/2610.09763