An Empirical Study of Metrics to Measure Representational Harms in Pre-Trained Language Models

By Saghar Hosseini, Hamid Palangi, Ahmed Hassan Awadallah. 2023

31 citations recorded by the source corpus.

Large-scale Pre-Trained Language Models (PTLMs) capture knowledge from massive human-written data which contains latent societal biases and toxic contents. In this paper, we leverage the primary task of PTLMs, i.e., language modeling, and propose a new metric to quantify manifested implicit representational harms in PTLMs towards 13 marginalized demographics. Using this metric, we conducted an empirical analysis of 24 widely used PTLMs. Our analysis provides insights into the correlation between the proposed metric in this work and other related metrics for representational harm. We observe that our metric correlates with most of the gender-specific metrics in the literature. Through extensive experiments, we explore the connections between PTLMs architectures and representational harms across two dimensions: depth and width of the networks. We found that prioritizing depth over width, mitigates representational harms in some PTLMs. Our code and data can be found at [place holder].

Subfields: Artificial Intelligence, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Safety, Robustness & Interpretability

Citations in the corpus

Llama 2: Open Foundation and Fine-Tuned Chat Models (2023, methodology): “We adopt a revised version of the dataset from Hosseini et al. (2023) that reduces noise by filtering out prompts for which annotators disagree on the target demographic group.”
Textbooks Are All You Need II: phi-1.5 technical report (2023, background): “In accordance with [HPA23], a subset of 6541 sentences are selected and scored based on scaled perplexity and sentence toxicity.”