skip to main content


Title: Quantifying Social Biases Using Templates is Unreliable
Recently, there has been an increase in efforts to understand how large language models (LLMs) propagate and amplify social biases. Several works have utilized templates for fairness evaluation, which allow researchers to quantify social biases in the absence of test sets with protected attribute labels. While template evaluation can be a convenient and helpful diagnostic tool to understand model deficiencies, it often uses a simplistic and limited set of templates. In this paper, we study whether bias measurements are sensitive to the choice of templates used for benchmarking. Specifically, we investigate the instability of bias measurements by manually modifying templates proposed in previous works in a semantically-preserving manner and measuring bias across these modifications. We find that bias values and resulting conclusions vary considerably across template modifications on four tasks, ranging from an 81% reduction (NLI) to a 162% increase (MLM) in (task-specific) bias measurements. Our results indicate that quantifying fairness in LLMs, as done in current practice, can be brittle and needs to be approached with more care and caution.  more » « less
Award ID(s):
2046873
NSF-PAR ID:
10462497
Author(s) / Creator(s):
; ;
Date Published:
Journal Name:
NeurIPS Workshop on Trustworthy and Socially Responsible Machine Learning (TSRML)
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Recent studies show that Natural Language Processing (NLP) technologies propagate societal biases about demographic groups associated with attributes such as gender, race, and nationality. To create interventions and mitigate these biases and associated harms, it is vital to be able to detect and measure such biases. While existing works propose bias evaluation and mitigation methods for various tasks, there remains a need to cohesively understand the biases and the specific harms they measure, and how different measures compare with each other. To address this gap, this work presents a practical framework of harms and a series of questions that practitioners can answer to guide the development of bias measures. As a validation of our framework and documentation questions, we also present several case studies of how existing bias measures in NLP—both intrinsic measures of bias in representations and extrinsic measures of bias of downstream applications—can be aligned with different harms and how our proposed documentation questions facilitates more holistic understanding of what bias measures are measuring. 
    more » « less
  2. AI systems have been known to amplify biases in real-world data. Explanations may help human-AI teams address these biases for fairer decision-making. Typically, explanations focus on salient input features. If a model is biased against some protected group, explanations may include features that demonstrate this bias, but when biases are realized through proxy features, the relationship between this proxy feature and the protected one may be less clear to a human. In this work, we study the effect of the presence of protected and proxy features on participants’ perception of model fairness and their ability to improve demographic parity over an AI alone. Further, we examine how different treatments—explanations, model bias disclosure and proxy correlation disclosure—affect fairness perception and parity. We find that explanations help people detect direct but not indirect biases. Additionally, regardless of bias type, explanations tend to increase agreement with model biases. Disclosures can help mitigate this effect for indirect biases, improving both unfairness recognition and decision-making fairness. We hope that our findings can help guide further research into advancing explanations in support of fair human-AI decision-making. 
    more » « less
  3. Attention filters sensory inputs to enhance task-relevant information. It is guided by an “attentional template” that represents the stimulus features that are currently relevant. To understand how the brain learns and uses templates, we trained monkeys to perform a visual search task that required them to repeatedly learn new attentional templates. Neural recordings found that templates were represented across the prefrontal and parietal cortex in a structured manner, such that perceptually neighboring templates had similar neural representations. When the task changed, a new attentional template was learned by incrementally shifting the template toward rewarded features. Finally, we found that attentional templates transformed stimulus features into a common value representation that allowed the same decision-making mechanisms to deploy attention, regardless of the identity of the template. Altogether, our results provide insight into the neural mechanisms by which the brain learns to control attention and how attention can be flexibly deployed across tasks. 
    more » « less
  4. With the rise of AI, algorithms have become better at learning underlying patterns from the training data including ingrained social biases based on gender, race, etc. Deployment of such algorithms to domains such as hiring, healthcare, law enforcement, etc. has raised serious concerns about fairness, accountability, trust and interpretability in machine learning algorithms. To alleviate this problem, we propose D-BIAS, a visual interactive tool that embodies human-in-the-loop AI approach for auditing and mitigating social biases from tabular datasets. It uses a graphical causal model to represent causal relationships among different features in the dataset and as a medium to inject domain knowledge. A user can detect the presence of bias against a group, say females, or a subgroup, say black females, by identifying unfair causal relationships in the causal network and using an array of fairness metrics. Thereafter, the user can mitigate bias by refining the causal model and acting on the unfair causal edges. For each interaction, say weakening/deleting a biased causal edge, the system uses a novel method to simulate a new (debiased) dataset based on the current causal model while ensuring a minimal change from the original dataset. Users can visually assess the impact of their interactions on different fairness metrics, utility metrics, data distortion, and the underlying data distribution. Once satisfied, they can download the debiased dataset and use it for any downstream application for fairer predictions. We evaluate D-BIAS by conducting experiments on 3 datasets and also a formal user study. We found that D-BIAS helps reduce bias significantly compared to the baseline debiasing approach across different fairness metrics while incurring little data distortion and a small loss in utility. Moreover, our human-in-the-loop based approach significantly outperforms an automated approach on trust, interpretability and accountability. 
    more » « less
  5. Charles, Cyril (Ed.)
    Manually collecting landmarks for quantifying complex morphological phenotypes can be laborious and subject to intra and interobserver errors. However, most automated landmarking methods for efficiency and consistency fall short of landmarking highly variable samples due to the bias introduced by the use of a single template. We introduce a fast and open source automated landmarking pipeline (MALPACA) that utilizes multiple templates for accommodating large-scale variations. We also introduce a K-means method of choosing the templates that can be used in conjunction with MALPACA, when no prior information for selecting templates is available. Our results confirm that MALPACA significantly outperforms single-template methods in landmarking both single and multi-species samples. K-means based template selection can also avoid choosing the worst set of templates when compared to random template selection. We further offer an example of post-hoc quality check for each individual template for further refinement. In summary, MALPACA is an efficient and reproducible method that can accommodate large morphological variability, such as those commonly found in evolutionary studies. To support the research community, we have developed open-source and user-friendly software tools for performing K-means multi-templates selection and MALPACA. 
    more » « less