Gorjan

SteeringTokens. Repository for the ACL 2026 paper "Compositional Steering of Large Language Models with Steering Tokens": https://arxiv.org/pdf/2601.05062

Synergy. Python

MLB. ◉ Improving Multimodal Learning with Multi-Loss Gradient Modulation (BMVC 2024) ◉ Self-Balancing Multimodal Models via Multi-Loss Gradient Modulation (IJCV 2025)

CoRe-Sleep. ◉ CoRe-Sleep: A Multimodal Fusion Framework for Time Series Robust to Imperfect Modalities (TNSRE 2024)

nanocode. Minimal Claude Code alternative. Single Python file, zero dependencies, ~250 lines.

SpecGloss-GS. [WACV 2026 Oral] Spec-Gloss Surfels and Normal–Diffuse Priors for Relightable Glossy Objects

nanochat. The best ChatGPT that $100 can buy.

MCR. ◉ Balancing Multimodal Training Through Game-Theoretic Regularization (Spotlight NeurIPS 2025)

ocebo. Object-Centric Pretraining via Target Encoder Bootstrapping (ICLR 2025)

ai-deadlines. ⏰ AI conference deadline countdowns

simzss. This repo contains the code for the ICLR 2025 paper: "A Simple Framework for Open-Vocabulary Zero-Shot Segmentation".

label-shift-calibration. Codebase for "LaSCal: Label-Shift Calibration without target labels", published at NeurIPS 2024.

aisuite. Simple, unified interface to multiple Generative AI providers

Bay-CAT. [ECCV’24] Official Implementation for CAT: Enhancing Multimodal Large Language Model to Answer Questions in Dynamic Audio-Visual Scenarios

proper-calibration-error. Codebase for "Consistent and Asymptotically Unbiased Estimation of Proper Calibration Errors", published at AISTATS 2024.

prismatic-vlms. A flexible and efficient codebase for training visually-conditioned language models (VLMs)

chug. Minimal sharded dataset loaders, decoders, and utils for multi-modal document, image, and text datasets.

CrIBo. This repo contains the code for the ICLR 2024 paper: "CrIBo: Self-Supervised Learning via Cross-Image Object-Level Bootstrapping".

Babel-ImageNet. Jupyter Notebook

tensordict. TensorDict is a pytorch dedicated tensor container.

calibration-object-detection. Codebase for "Beyond Classification: Definition and Density-based Estimation of Calibration in Object Detection", published at WACV 2024.

WEHI-ResearchComputing.github.io. This is the website for the RCP at WEHI

abstention. Algorithms for abstention, calibration and domain adaptation to label shift.

SynthIE. The data and the PyTorch implementation for the models and experiments in the paper "Exploiting Asymmetry for Synthetic Training Data Generation: SynthIE and the Case of Information Extraction".

fact-linking. Codebase for "Linking Surface Facts to Large-Scale Knowledge Graphs" (EMNLP 2023)

AdaSim. This repo contains the code for the ICCV 2023 paper: "Adaptive Similarity Bootstrapping for Self-Distillation based Representation Learning".

global-local-self-distillation. This repo contains the code for the WACV 2023 paper Global-Local Self-Distillation for Visual Representation Learning.

motion-prediction-tim. This repo contains the code for the ACCV 2020 paper Motion Prediction using Temporal Inception Module.

CrOC. This repo contains the code for the CVPR 2023 paper: "CrOC : Cross-View Online Clustering for Dense Visual Representation Learning".

mBLIP. Python

indexed_gzip. Fast random access of gzip files in Python

wikimapper. Mapping Wikipedia pages to Wikidata IDs and vice versa.

faiss. A library for efficient similarity search and clustering of dense vectors.

confidence-intervals. Official code repository for "Distribution-Independent Confidence Intervals for the Eigendecomposition of Covariance Matrices via the Eigenvalue-Eigenvector Identity" (ICML 2021 Workshop on Distribution-Free Uncertainty Quantification).

explanatory-guided-learning. Codebase for “Machine Guides, Human Supervises: Interactive Learning with Global Explanations”.

ece-kde. Codebase for "A Consistent and Differentiable Lp Canonical Calibration Error Estimator", published at NeurIPS 2022.

revisiting-spatial-temporal-layouts. Codebase for "Revisiting spatio-temporal layouts for compositional action recognition" (Oral at BMVC 2021).

pixel. Research code for pixel-based encoders of language (PIXEL)

rezero. Official PyTorch Repo for "ReZero is All You Need: Fast Convergence at Large Depth"

refer. Referring Expression Datasets API

MTCN. Implementation of "With a Little Help from my Temporal Context: Multimodal Egocentric Action Recognition, BMVC, 2021" in PyTorch

scene_graph_benchmark. image scene graph generation benchmark

startjekyll. An example and guide to getting started with Jekyll and static site generators.

ConvNeXt. Code release for ConvNeXt model

img2dataset. Easily turn large sets of image urls to an image dataset. Can download, resize and package 100M urls in 20h on one machine.

calibration_and_bias. Codebase for "On the relationship between calibrated predictors and unbiased volume estimation" (MICCAI 2021).

clever. The Curious Layperson: Fine-Grained Image Recognition without Expert Labels (BMVC 2021 best student paper)

SimCSE. [EMNLP 2021] SimCSE: Simple Contrastive Learning of Sentence Embeddings https://arxiv.org/abs/2104.08821

bertviz. BertViz: Visualize Attention in Transformer Models

perceiver-pytorch. Implementation of Perceiver, General Perception with Iterative Attention, in Pytorch

x-transformers. A concise but complete full-attention transformer with a set of promising experimental features from various papers

Multimodal-Alignment-Framework. Implementation for MAF: Multimodal Alignment Framework

paperswithcode-client. API Client for paperswithcode.com

einops. Flexible and powerful tensor operations for readable and reliable code (for pytorch, jax, TF and others)

do-you-even-need-attention. Is the attention layer even necessary? (https://arxiv.org/abs/2105.02723)

SceneGraphParser. A python toolkit for parsing captions (in natural language) into scene graphs (as symbolic representations).

accelerate. 🚀 A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (including fp8), and easy-to-configure FSDP and DeepSpeed support

torchtyping. Type annotations and dynamic checking for a tensor's shape, dtype, names, etc.

notation. TeX

3D-ResNets-PyTorch. 3D ResNets for Action Recognition (CVPR 2018)

TRN-pytorch. Temporal Relation Networks

glances. Glances an Eye on your system. A top/htop alternative for GNU/Linux, BSD, Mac OS and Windows operating systems.

MiDaS. Code for robust monocular depth estimation described in "Ranftl et. al., Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer, TPAMI 2022"

no_frills_hoi_det. A strong HOI Detection model without Frills!

detectron2. Detectron2 is a platform for object detection, segmentation and other visual recognition tasks.

detr. End-to-End Object Detection with Transformers

ffmpeg-python. Python bindings for FFmpeg - with complex filtering support

vit-pytorch. Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch

S3D. Release of the pretrained S3D Network in PyTorch (ECCV 2018)

something_else. Code repository for the paper: 'Something-Else: Compositional Action Recognition with Spatial-Temporal Interaction Networks'

awesome-shell. A curated list of awesome command-line frameworks, toolkits, guides and gizmos. Inspired by awesome-php.

exa. A modern replacement for ‘ls’.

minGPT. A minimal PyTorch re-implementation of the OpenAI GPT (Generative Pretrained Transformer) training

PyTorch-GAN. PyTorch implementations of Generative Adversarial Networks.

latexify_py. A library to generate LaTeX expression from Python code.

localized-narratives. Localized Narratives

dlbook_exercises. Exercises for the Deep Learning textbook at www.deeplearningbook.org

papers-with-annotations. Research papers with annotations, illustrations and explanations

pytorch-image-models. The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNetV4, MobileNet-V3 & V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeXt, and more

datasets. 🤗 The largest hub of ready-to-use datasets for AI models with fast, easy-to-use and efficient data manipulation tools

open_lth. A repository in preparation for open-sourcing lottery ticket hypothesis code.

wttr.in. :partly_sunny: The right way to check the weather

git-lfs. Git extension for versioning large files

awesome-papers. Papers & presentation materials from Hugging Face's internal science day

gdown. Google Drive public file downloader when curl/wget fails.

NYU-DLSP20. NYU Deep Learning Spring 2020

torchlayers. Shape and dimension inference (Keras-like) for PyTorch layers and neural networks

d2l-en. Interactive deep learning book with multi-framework code, math, and discussions. Adopted at 500 universities from 70 countries including Stanford, MIT, Harvard, and Cambridge.

voila. Voilà turns Jupyter notebooks into standalone web applications

virus-spreading. Simple virus spreading simulation tool made with plain/vanilla JavaScript

solt. Streaming over lightweight data transformations

Made-With-ML. Learn how to develop, deploy and iterate on production-grade ML applications.

knockknock. 🚪✊Knock Knock: Get notified when your training ends with only two additional lines of code

fastbook. The fastai book, published as Jupyter Notebooks

bat. A cat(1) clone with wings.

jax. Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

pytorch-metric-learning. The easiest way to use deep metric learning in your application. Modular, flexible, and extensible. Written in PyTorch.

conceptual-captions. Conceptual Captions is a dataset containing (image-URL, caption) pairs designed for the training and evaluation of machine learned image captioning systems.

brave-browser. Brave browser for Android, iOS, Linux, macOS, Windows.

vscode. Visual Studio Code