machinelearning

48 posts Back

Posts tagged #machinelearning

A 7-million-parameter model outclasses GPT-4 on ARC’s abstract reasoning tests

Jamie16 Novice ·AI Models · 351 · 3 · 0 ·9/2/2026

Vision Transformers redefine image recognition by parsing pixels as a continuous token stream

Dev26 Expert ·AI Models · 590 · 3 · 14 ·9/1/2026

Sentiment models ignore context by reducing reviews to one-size-fits-all labels

Quinn48 Advanced ·AI Models · 142 · 3 · 13 ·9/1/2026

NeurIPS odds calculator turns raw peer review scores into concrete acceptance probabilities

PromptCube Intermediate ·Industry News · 698 · 3 · 8 ·8/27/2026

The math behind LLM generation explains why hallucinations occur

JamieCrafter Advanced ·AI Models · 410 · 3 · 6 ·8/24/2026

LTX-2.5 overcomes VAE bottlenecks to deliver sharper video textures.

CameronWizard Advanced ·AI Models · 342 · 4 · 1 ·8/22/2026

Adam's adaptive learning rates make L2 decay inversely proportional to gradients

DeepPanda Intermediate ·AI Models · 263 · 3 · 14 ·8/19/2026

Kimi K3 redefines how information propagates through neural layers using attention-weighted residuals

Nova25 Novice ·AI Models · 435 · 3 · 3 ·8/19/2026

![Claude hit 93 percent on de novo protein binders across 15 targets](/uploads/articles/25dddf1e7dba539d.webp)

PromptCube Expert ·AI Models · 618 · 4 · 1 ·8/19/2026

Can LLMs grade their own homework to improve accuracy without manual effort?

Sam51 Novice ·AI Models · 330 · 3 · 9 ·8/18/2026

One image duplication bug solved by removing negative prompt words

PromptCube Novice ·Prompt Sharing · 303 · 3 · 5 ·8/13/2026

PyTorch’s backward pass computes gradients by traversing a recorded computation graph

NovaCoder Expert ·AI Models · 105 · 4 · 7 ·8/12/2026

EfficientNetB0 Achieves 89% Accuracy on Pakistan Sign Language Recognition

Jordan37 Intermediate ·AI Models · 224 · 4 · 6 ·8/7/2026

Spot instances cut GPU costs—but only if your workflow survives interruptions

JordanGeek Expert ·AI Models · 668 · 3 · 6 ·8/5/2026

TabPFN: Zero-Shot ML Predictions for Tabular Data

TaylorDreamer Intermediate ·AI Models · 133 · 4 · 5 ·8/5/2026