推荐/广告/搜索领域工业界经典以及最前沿论文集合。A collection of industry classics and cutting-edge papers in the field of recommendation/advertising/search.

Remove problematic gender bias from word embeddings.

Counterfactual Samples Synthesizing for Robust VQA

[ICCV 2023] PyTorch implementation of RandBox

[EMNLP 2023] Official Code of "JointMatch: A Unified Approach for Diverse and Collaborative Pseudo-Labeling to Semi-Supervised Text Classification"

Debiasing both contextual and context free embeddings and developing a downstream metric for bias measurement

This repository is a debiassing project on CXR Dataset with embedding which is publicly available.

[EMNLP 2023] Official Code of "DeCrisisMB: Debiased Semi-Supervised Learning for Crisis Tweet Classification via Memory Bank"

Course 5 from deeplearning.ai

Extension of bias-bench for select Sentence Transformers* models

nexttext is a lightweight Python toolkit for neutralizing explicit gendered terms in text. Via simple, fast heuristics it helps clean and standardize AI-generated labels, reducing model-induced misgendering and other unwanted linguistic artifacts. It's intended for use in AI pipelines, for both processing training data or filtering model outputs.

Identifying gender bias in film reviews through three distinct Natural Language Processing models