Understanding the Effect of Model Compression on Social Bias in Large Language Models

2023/12/09 by Gustavo Gonçalves, Emma Strubell, Gonçalves, Gustavo +1 · 8 citations
Computer Science · #Computation and Language (cs.CL) #Explainable Artificial Intelligence (XAI) #FOS: Computer and information sciences #Natural Language Processing Techniques #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2312.05662

openalex publication_date 2023/12/09 · openalex created_date 2023/12/14 · openalex updated_date 2026/07/28

Abstract

Large Language Models (LLMs) trained with self-supervision on vast corpora of web text fit to the social biases of that text. Without intervention, these social biases persist in the model's predictions in downstream tasks, leading to representational harm. Many strategies have been proposed to mitigate the effects of inappropriate social biases learned during pretraining. Simultaneously, methods for model compression have become increasingly popular to reduce the computational burden of LLMs. Despite the popularity and need for both approaches, little work has been done to explore the interplay between these two. We perform a carefully controlled study of the impact of model compression via quantization and knowledge distillation on measures of social bias in LLMs. Longer pretraining and larger models led to higher social bias, and quantization showed a regularizer effect with its best trade-off around 20% of the original pretraining time.

Cited by

• Downsized and Compromised?: Assessing the Faithfulness of Model Compression
• Debiasing Large Language Models in Thai Political Stance Detection via Counterfactual Calibration
• Fair-GPTQ: Bias-Aware Quantization for Large Language Models
• How Quantization Shapes Bias in Large Language Models
• Investigating Hallucination in Conversations for Low Resource Languages
• WINA: Weight Informed Neuron Activation for Accelerating Large Language Model Inference
• Safety Alignment via Constrained Knowledge Unlearning
• Quantization Undoes Alignment: Bias Emergence in Compressed LLMs Across Models and Precision Levels

Related

• Social Bias Evaluation for Large Language Models Requires Prompt Variations
• A Predictive Factor Analysis of Social Biases and Task-Performance in Pretrained Masked Language Models
• Explicit vs. Implicit: Investigating Social Bias in Large Language Models through Self-Reflection
• Evaluation of Social Biases in Recent Large Pre-Trained Models
• Understanding the Interplay of Scale, Data, and Bias in Language Models: A Case Study with BERT