Machine Learning

The latest machine learning stories, curated continuously from hundreds of sources.

A Neural Network Learned to Read Digits From Pure Noise

Machine LearningTowards Data ScienceUtkarsh Mangal5h

Are there fundamental repeating patterns in ML techniques across modalities (text, image, audio) that let us group them? [R]

Machine LearningReddit r/MachineLearning/u/OkCampaign840017h

Is neuro-evolution relevant? [D]

Machine LearningReddit r/MachineLearning/u/tapopalis23h

Hierarchical detection of neonatal brain pathologies using a two-stage deep learning framework from magnetic resonance imaging [R]

Machine LearningReddit r/MachineLearning/u/rottoneuro1d

First-time conference attendee at AACL-IJCNLP 2026 (Hengqin), going solo, looking for guidance and people to meet

Machine LearningReddit r/LanguageTechnology/u/Nearby_Reaction29471d

Tested whether AI noise cleaning improves transcription accuracy for Hindi-English audio. It makes it worse.

Machine LearningReddit r/LanguageTechnology/u/Noisecleanai1d

Multi-Token Prediction Done Right [R]

Machine LearningReddit r/MachineLearning/u/jayden_teoh_1d

nemotron-halloween

Machine LearningHugging Face Blog1d

Making it easier for websites to express AI training preferences

Machine LearningOpenAI News2d

Investigating the Early Signatures of Feature Learning in Deep Linear Networks

Machine LearningarXiv stat.MLShaunak Bhandarkar, Boris…2d

Predicate Neuron: Differentiable Quantization of Predicate Relations in Gradient-based Learning

Machine LearningarXiv cs.LGIoannis Konstantoulas, Il…2d

Deep learning for non-transvenous ICD screening: tailored neural network architectures, extended experiments, and explainability

Machine LearningarXiv cs.LGBartosz Morawski, Samuel…2d

A Separation Principle for Adversarial Attacks and Defense of Graph Neural Networks

Machine LearningarXiv cs.LGFarid Mafi, Saber Jafarpo…2d

Replaying the Same Gradients at a Different $\varepsilon$: A Registered Test of the Fixed-Trajectory Account of Adam's $\varepsilon$ for One Rare Row

Machine LearningarXiv cs.LGLngel Kyo (A2O Labs)2d

A Gradient based Adversarial Augmentation Framework for Mitigating Unlearnable Datasets

Machine LearningarXiv cs.LGLong Dang, Satpal Sura, K…2d

Moment-Guided Optimizers for Noisy Deep Learning: MGO-SC as a Simple Consistency Gate for Adam-Style Updates

Machine LearningarXiv cs.LGAlicem Koyun2d

Sometimes less is more: Investigating inductive biases of MLPs and Transformers on noisy k-parity features

Machine LearningarXiv cs.LGAlasdair Paren2d

Fair use benefits everyone

Machine LearningOpenAI NewsOct 30

Granite Switch 4.1.1 — Update models like you update software

Machine LearningHugging Face Blog3d

Announcing Project ProtoStar: Making Publicly Funded Science AI-Ready

Machine LearningHugging Face BlogOct 29

Welcome Le Chonk: Mistral Large 4 on Hugging Face

Machine LearningHugging Face BlogOct 29

Introducing AnySpeechRecognizer and AnyDecisionModel: Open Models for Speech and Decisions on Apple Platforms

Machine LearningHugging Face BlogOct 29

How we’re training a Figma-fluent editing model

Machine LearningFigma Engineering BlogOct 29

CPT’d Qwen9B on Polish legal data. How to make it chat? IMPORTANTLY: how to make it reason? and function call in my harness? and which to do first? [D]

Machine LearningReddit r/MachineLearning/u/SignificantZebra58833d

Running NVIDIA's LocateAnything-3B (open-vocabulary detection) fully locally from .NET, on a live camera stream

Machine LearningReddit r/computervision/u/MasterOfAutomation3d

[Anomaly Detection] PatchCore on Orange line 🍊

Machine LearningReddit r/computervision/u/erik_kokalj3d

Single-camera padel analysis with RF-DETR, TrackNet V3 and audio–visual hit detection

Machine LearningReddit r/computervision/u/ResponsibilityFair333d

Distributed training with Ray Train DLCs and EFA on EKS

Machine LearningAWS Machine Learning BlogSunita Koppar3d

Is there a good way to automatically skip unwanted

Machine LearningReddit r/computervision/u/Cool-Bid62633d

I am building a $67 camera module that does SLAM and VPS onboard

Machine LearningReddit r/computervision/u/twokiloballs4d

Video to 3D

Machine LearningReddit r/computervision/u/Mohamed_1834d

Nvidia's open Nemotron 3.5 Transcribe combines transcription, translation, and speaker recognition

Machine LearningThe DecoderJonathan Kemper4d

NVIDIA Nemotron 3.5 Transcribe and Granary v2: Intelligent Multilingual Transcription You Can Prompt and Adapt

Machine LearningHugging Face Blog4d

got this recording and have no idea what they’re saying lol

Machine LearningReddit r/LanguageTechnology/u/JobPrior78254d

Pandora M1, a training monitor box (64 nodes x 8 GPUs support) [P]

Machine LearningReddit r/MachineLearning/u/perone4d

Computational and Statistical Guarantees of the c-Rectified flow

Machine LearningarXiv stat.MLLeda Wang, Zhehao Xu, Qia…4d

Scaling and Renormalization in Neural Networks: An Effective Renormalization-Group Description of Grokking

Machine LearningarXiv cs.LGMin Chen, Junqi Wang, Zeg…4d

Multiple Leak Localization in Water Distribution Networks through Graph Neural Networks

Machine LearningarXiv cs.LGArvin Ajoodani, Sara Nazi…4d

Dynamical coarse-graining of neural-network training

Machine LearningarXiv cs.LGLin-Juan Wang, Da Li, Hui…4d

Repulsive Population Policy Gradients for Behaviorally Diverse Reinforcement Learning

Machine LearningarXiv cs.LGSaeed Hedayatian, Harsh S…4d