paper-with-me

홈 › Papers

A Trip Towards Fairness: Bias and De-Biasing in Large Language Models

2023-05-23 · Leonardo Ranaldi, Elena Sofia Ruzzetti, Davide Venditti, Dario Onorati, Fabio Massimo Zanzotto

Cheap-to-Build Very Large-Language Models (CtB-LLMs) with affordable training are emerging as the next big revolution in natural language processing and understanding. These CtB-LLMs are democratizing access to trainable Very Large-Language Models (VLLMs) and, thus, may represent the building blocks of many NLP systems solving downstream tasks. Hence, a little or a large bias in CtB-LLMs may cause huge harm. In this paper, we performed a large investigation of the bias of three families of CtB-LLMs, and we showed that debiasing techniques are effective and usable. Indeed, according to current tests, the LLaMA and the OPT families have an important bias in gender, race, religion, and profession. In contrast to the analysis for other LLMs, we discovered that bias depends not on the number of parameters but on the perplexity. Finally, the debiasing of OPT using LoRA reduces bias up to 4.12 points in the normalized stereotype score.

📄 PDF Abstract BibTeX arXiv:2305.13862

Code (0)

등록된 구현이 없습니다.

Tasks

Fairness

Methods 이 논문이 사용한 방법론

Attention 설명 없음
OPT OPT is a suite of decoder-only pre-trained transformers ranging from 125M to 175B parameters. The model uses an AdamW optimizer and weight decay of 0.1. It follows a linear…
Cosine Annealing Cosine Annealing is a type of learning rate schedule that has the effect of starting with a large learning rate that is relatively rapidly decreased to a minimum value before…
Weight Decay 설명 없음
Residual Connection 설명 없음
Linear Warmup With Cosine Annealing Linear Warmup With Cosine Annealing is a learning rate schedule where we increase the learning rate linearly for $n$ updates and then anneal according to a cosine schedule…
Refunds@Expedia|||How do I get a full refund from Expedia? “How do I get a full refund from Expedia? How do I get a full refund from Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Quick Help &…
Discriminative Fine-Tuning Discriminative Fine-Tuning is a fine-tuning strategy that is used for ULMFiT type models. Instead of using the same learning rate…

Similar Papers 제목 키워드 기반

KLAAD: Refining Attention Mechanisms to Reduce Societal Bias in Generative Language Models

2025-07-26 · Seorin Kim, Dongyoung Lee, Jaejin Lee arxiv

Large language models (LLMs) often exhibit societal biases in their outputs, prompting ethical concerns regarding fairness and harm. In this work, we propose KLAAD (KL-Attention Alignment Debiasing), an attention-based d…

Large Language Model Bias Mitigation from the Perspective of Knowledge Editing

2024-05-15 · Ruizhe Chen, Yichen Li, Zikai Xiao, Zuozhu Liu

Existing debiasing methods inevitably make unreasonable or undesired predictions as they are designated and evaluated to achieve parity across different social groups but leave aside individual facts, resulting in modifi…

Fairnessknowledge editingLanguage ModelingLanguage Modelling+2

BiasFilter: An Inference-Time Debiasing Framework for Large Language Models

2025-05-28 · Xiaoqing Cheng, Ruizhe Chen, Hongying Zan, Yuxiang Jia 외

Mitigating social bias in large language models (LLMs) has become an increasingly important research objective. However, existing debiasing methods often incur high human and computational costs, exhibit limited effectiv…

Fairness

A Survey on Fairness in Large Language Models

2023-08-20 · Yingji Li, Mengnan Du, Rui Song, Xin Wang 외

Large Language Models (LLMs) have shown powerful performance and development prospects and are widely deployed in the real world. However, LLMs can capture social biases from unprocessed training data and propagate the b…

FairnessSurvey

Selective Test-Time Debiasing for CLIP via Reward Gating

2026-07-01 · Jaeho Han, Jisoo Yang, Hyeondong Woo, Mingyu Jeon 외 arxiv

Vision language models (VLMs) demonstrate strong zero-shot performance, but often perpetuate social stereotypes in person-centric queries, yielding skewed demographic distributions. Current debiasing methods apply unifor…

Reinforcement LearningTest-time Adaptation