🔥🔥High-Performance Face Recognition Library on PaddlePaddle & PyTorch🔥🔥

[ICML 2022] "ProGCL: Rethinking Hard Negative Mining in Graph Contrastive Learning"

Robust Contrastive Learning Using Negative Samples with Diminished Semantics (NeurIPS 2021)

Code and data of the EMNLP 2022 Main Conference paper "Reduce Catastrophic Forgetting of Dense Retrieval Training with Teleportation Negatives".

Detection algorithms and applications from famous papers; simple theory; solid code.

Official code for TNNLS paper "Affinity Uncertainty-based Hard Negative Mining in Graph Contrastive Learning"

https://teacher.bupt.edu.cn/zhuchuang/en/index.htm

Train a SVM and for detecting human upper bodies in TV series The Big Bang Theory

ViIR: The Unified Framework for Fine-tuning Vietnamese Information Retrieval Models with Various Tuning Statergies.

SuperGlue training pipeline written using Pytorch Lightning

Winner project of the HackUDC 2026 Inditex AI challenge. From a photo of an outfit and given the Zara catalog, identify all the individual pieces the model is wearing.

Custom Object Detector

Implements relation extraction for biomedical texts using Hard Negative Mining to improve accuracy in identifying complex entity relationships. Includes code for data processing, training, and evaluation with BioC-format datasets.

Spring 2022 Bioimage Informatics (Self-Study ) project using triplet loss and hard negative mining

Object detector architecture research: compares YOLOv8m and RetinaNet+MiT-b0 on a custom-annotated ecology dataset with hard-negative mining and explainable AI galleries. Ablations include optimiser family, learning rate scheduling, progressive unfreezing and augmentations.

This repository stores work for a seminar for my master's degree. The course is held from the chair Intelligent Embedded Systems (IES). My topic is Approaches for finding sample pairs in contrastive learning.