FairPy: A Toolkit for Evaluation of Prediction Biases and their Mitigation in Large Language Models

• URL: http://arxiv.org/abs/2302.05508v2
• Date: Tue, 15 Apr 2025 04:08:13 GMT
• Title: FairPy: A Toolkit for Evaluation of Prediction Biases and their Mitigation in Large Language Models
• Authors: Hrishikesh Viswanath, Tianyi Zhang,
• Abstract summary: Recent studies have demonstrated that large pretrained language models (LLMs) such as BERT and GPT-2 exhibit biases in token prediction.<n>We present a comprehensive survey of such techniques tailored towards widely used LLMs such as BERT, GPT-2, etc.<n>We additionally introduce Fairpy, a modular and toolkit that provides plug-and-play interfaces for integrating these mathematical tools.
• Score: 12.62204775625353
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent studies have demonstrated that large pretrained language models (LLMs) such as BERT and GPT-2 exhibit biases in token prediction, often inherited from the data distributions present in their training corpora. In response, a number of mathematical frameworks have been proposed to quantify, identify, and mitigate these the likelihood of biased token predictions. In this paper, we present a comprehensive survey of such techniques tailored towards widely used LLMs such as BERT, GPT-2, etc. We additionally introduce Fairpy, a modular and extensible toolkit that provides plug-and-play interfaces for integrating these mathematical tools, enabling users to evaluate both pretrained and custom language models. Fairpy supports the implementation of existing debiasing algorithms. The toolkit is open-source and publicly available at: \href{https://github.com/HrishikeshVish/Fairpy}{https://github.com/HrishikeshVish/Fairpy}

Related papers

• Scalable Influence and Fact Tracing for Large Language Model Pretraining [14.598556308631018]
  Training data attribution (TDA) methods aim to attribute model outputs back to specific training examples. We refine existing gradient-based methods to work effectively at scale. We release our prompt set and model outputs, along with a web-based visualization tool to explore influential examples.
  arXiv  Detail & Related papers  (2024-10-22T20:39:21Z)
• FairX: A comprehensive benchmarking tool for model analysis using fairness, utility, and explainability [4.1942958779358674]
  We present FairX, an open-source benchmarking tool designed for the comprehensive analysis of models under the umbrella of fairness, utility, and eXplainability (XAI) FairX enables users to train benchmarking bias-mitigation models and evaluate their fairness using a wide array of fairness metrics, data utility metrics, and generate explanations for model predictions, all within a unified framework.
  arXiv  Detail & Related papers  (2024-06-20T13:07:06Z)
• Quantifying Bias in Text-to-Image Generative Models [49.60774626839712]
  Bias in text-to-image (T2I) models can propagate unfair social representations and may be used to aggressively market ideas or push controversial agendas. Existing T2I model bias evaluation methods only focus on social biases. We propose an evaluation methodology to quantify general biases in T2I generative models, without any preconceived notions.
  arXiv  Detail & Related papers  (2023-12-20T14:26:54Z)
• Evaluating and Explaining Large Language Models for Code Using Syntactic Structures [74.93762031957883]
  This paper introduces ASTxplainer, an explainability method specific to Large Language Models for code. At its core, ASTxplainer provides an automated method for aligning token predictions with AST nodes. We perform an empirical evaluation on 12 popular LLMs for code using a curated dataset of the most popular GitHub projects.
  arXiv  Detail & Related papers  (2023-08-07T18:50:57Z)
• CBBQ: A Chinese Bias Benchmark Dataset Curated with Human-AI Collaboration for Large Language Models [52.25049362267279]
  We present a Chinese Bias Benchmark dataset that consists of over 100K questions jointly constructed by human experts and generative language models. The testing instances in the dataset are automatically derived from 3K+ high-quality templates manually authored with stringent quality control. Extensive experiments demonstrate the effectiveness of the dataset in detecting model bias, with all 10 publicly available Chinese large language models exhibiting strong bias in certain categories.
  arXiv  Detail & Related papers  (2023-06-28T14:14:44Z)
• Exposing Bias in Online Communities through Large-Scale Language Models [3.04585143845864]
  This work uses the flaw of bias in language models to explore the biases of six different online communities. The bias of the resulting models is evaluated by prompting the models with different demographics and comparing the sentiment and toxicity values of these generations. This work not only affirms how easily bias is absorbed from training data but also presents a scalable method to identify and compare the bias of different datasets or communities.
  arXiv  Detail & Related papers  (2023-06-04T08:09:26Z)
• Gender Biases in Automatic Evaluation Metrics for Image Captioning [87.15170977240643]
  We conduct a systematic study of gender biases in model-based evaluation metrics for image captioning tasks. We demonstrate the negative consequences of using these biased metrics, including the inability to differentiate between biased and unbiased generations. We present a simple and effective way to mitigate the metric bias without hurting the correlations with human judgments.
  arXiv  Detail & Related papers  (2023-05-24T04:27:40Z)
• Bipol: A Novel Multi-Axes Bias Evaluation Metric with Explainability for NLP [0.276240219662896]
  We introduce bipol, a new metric with explainability, for estimating social bias in text data. In a step to address this challenge we create a novel metric that involves a two-step process. We create a large dataset for training models in bias detection and make it publicly available.
  arXiv  Detail & Related papers  (2023-04-08T14:45:15Z)
• FineDeb: A Debiasing Framework for Language Models [3.7698299781999376]
  We propose FineDeb, a two-phase debiasing framework for language models. Our results show that FineDeb offers stronger debiasing in comparison to other methods. Our framework is generalizable for demographics with multiple classes.
  arXiv  Detail & Related papers  (2023-02-05T18:35:21Z)
• Debiasing Vision-Language Models via Biased Prompts [79.04467131711775]
  We propose a general approach for debiasing vision-language foundation models by projecting out biased directions in the text embedding. We show that debiasing only the text embedding with a calibrated projection matrix suffices to yield robust classifiers and fair generative models.
  arXiv  Detail & Related papers  (2023-01-31T20:09:33Z)
• Fairness and Accuracy under Domain Generalization [10.661409428935494]
  Concerns have arisen that machine learning algorithms may be biased against certain social groups. Many approaches have been proposed to make ML models fair, but they typically rely on the assumption that data distributions in training and deployment are identical. We study the transfer of both fairness and accuracy under domain generalization where the data at test time may be sampled from never-before-seen domains.
  arXiv  Detail & Related papers  (2023-01-30T23:10:17Z)
• An Analysis of Social Biases Present in BERT Variants Across Multiple Languages [0.0]
  We investigate the bias present in monolingual BERT models across a diverse set of languages. We propose a template-based method to measure any kind of bias, based on sentence pseudo-likelihood. We conclude that current methods of probing for bias are highly language-dependent.
  arXiv  Detail & Related papers  (2022-11-25T23:38:08Z)
• MRCLens: an MRC Dataset Bias Detection Toolkit [82.44296974850639]
  We introduce MRCLens, a toolkit that detects whether biases exist before users train the full model. For the convenience of introducing the toolkit, we also provide a categorization of common biases in MRC.
  arXiv  Detail & Related papers  (2022-07-18T21:05:39Z)
• Challenges in Measuring Bias via Open-Ended Language Generation [1.5552869983952944]
  We analyze how specific choices of prompt sets, metrics, automatic tools and sampling strategies affect bias results. We provide recommendations for reporting biases in open-ended language generation.
  arXiv  Detail & Related papers  (2022-05-23T19:57:15Z)
• "I'm sorry to hear that": Finding New Biases in Language Models with a Holistic Descriptor Dataset [12.000335510088648]
  We present a new, more inclusive bias measurement dataset, HolisticBias, which includes nearly 600 descriptor terms across 13 different demographic axes. HolisticBias was assembled in a participatory process including experts and community members with lived experience of these terms. We demonstrate that HolisticBias is effective at measuring previously undetectable biases in token likelihoods from language models.
  arXiv  Detail & Related papers  (2022-05-18T20:37:25Z)
• Efficient Nearest Neighbor Language Models [114.40866461741795]
  Non-parametric neural language models (NLMs) learn predictive distributions of text utilizing an external datastore. We show how to achieve up to a 6x speed-up in inference speed while retaining comparable performance.
  arXiv  Detail & Related papers  (2021-09-09T12:32:28Z)
• Combining Feature and Instance Attribution to Detect Artifacts [62.63504976810927]
  We propose methods to facilitate identification of training data artifacts. We show that this proposed training-feature attribution approach can be used to uncover artifacts in training data. We execute a small user study to evaluate whether these methods are useful to NLP researchers in practice.
  arXiv  Detail & Related papers  (2021-07-01T09:26:13Z)
• Pre-training Multilingual Neural Machine Translation by Leveraging Alignment Information [72.2412707779571]
  mRASP is an approach to pre-train a universal multilingual neural machine translation model. We carry out experiments on 42 translation directions across a diverse setting, including low, medium, rich resource, and as well as transferring to exotic language pairs.
  arXiv  Detail & Related papers  (2020-10-07T03:57:54Z)
• REVISE: A Tool for Measuring and Mitigating Bias in Visual Datasets [64.76453161039973]
  REVISE (REvealing VIsual biaSEs) is a tool that assists in the investigation of a visual dataset. It surfacing potential biases along three dimensions: (1) object-based, (2) person-based, and (3) geography-based.
  arXiv  Detail & Related papers  (2020-04-16T23:54:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.