paper-with-me

홈 › Papers

Controlling Gender Bias in Retrieval via a Backpack Architecture

2025-11-02 · Amirabbas Afzali, Amirreza Velae, Iman Ahmadi, Mohammad Aliannejadi arxiv

The presence of social biases in large language models (LLMs) has become a significant concern in AI research. These biases, often embedded in training data, can perpetuate harmful stereotypes and distort decision-making processes. When LLMs are integrated into ranking systems, they can propagate these biases, leading to unfair outcomes in critical applications such as search engines and recommendation systems. Backpack Language Models, unlike traditional transformer-based models that treat text sequences as monolithic structures, generate outputs as weighted combinations of non-contextual, learned word aspects, also known as senses. Leveraging this architecture, we propose a framework for debiasing ranking tasks. Our experimental results show that this framework effectively mitigates gender bias in text retrieval and ranking with minimal degradation in performance.

📄 PDF Abstract BibTeX arXiv:2511.00875

Code (0)

등록된 구현이 없습니다.

Tasks

Recommendation SystemsText Retrieval

Similar Papers 제목 키워드 기반

Backpack Language Models

2023-05-26 · John Hewitt, John Thickstun, Christopher D. Manning, Percy Liang

We present Backpacks: a new neural architecture that marries strong modeling performance with an interface for interpretability and control. Backpacks learn multiple non-contextual sense vectors for each word in a vocabu…

Language ModelingLanguage ModellingText GenerationWord Embeddings

Character-level Chinese Backpack Language Models

2023-10-19 · Hao Sun, John Hewitt

The Backpack is a Transformer alternative shown to improve interpretability in English language modeling by decomposing predictions into a weighted sum of token sense components. However, Backpacks' reliance on token-def…

Language ModelingLanguage Modelling

Mitigating Bias in RAG: Controlling the Embedder

2025-02-24 · Taeyoun Kim, Jacob Springer, aditi raghunathan, Maarten Sap

In retrieval augmented generation (RAG) systems, each individual component -- the LLM, embedder, and corpus -- could introduce biases in the form of skews towards outputting certain perspectives or identities. In this wo…

FairnessRAGRetrieval-augmented Generation

Measuring Gender Bias in Job Title Matching for Grammatical Gender Languages

2025-09-17 · Laura García-Sardiña, Hermenegildo Fabregat, Daniel Deniz, Rabih Zbib arxiv

This work sets the ground for studying how explicit grammatical gender assignment in job titles can affect the results of automatic job ranking systems. We propose the usage of metrics for ranking comparison controlling …

Multi-Dimensional Gender Bias Classification

2020-05-01 · EMNLP 2020 11 · Emily Dinan, Angela Fan, Ledell Wu, Jason Weston 외

Machine learning models are trained to find patterns in data. NLP models can inadvertently learn socially undesirable patterns when training on gender biased text. In this work, we propose a general framework that decomp…

ClassificationGeneral Classification