Controlling Gender Bias in Retrieval via a Backpack Architecture
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.
Code (0)
등록된 구현이 없습니다.
Tasks
Recommendation SystemsText RetrievalSimilar Papers 제목 키워드 기반
Backpack Language Models
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 EmbeddingsCharacter-level Chinese Backpack Language Models
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 ModellingMitigating Bias in RAG: Controlling the Embedder
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 GenerationMeasuring Gender Bias in Job Title Matching for Grammatical Gender Languages
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
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