Accurate and Data-Efficient Toxicity Prediction when Annotators Disagree

• URL: http://arxiv.org/abs/2410.12217v1
• Date: Wed, 16 Oct 2024 04:26:40 GMT
• Title: Accurate and Data-Efficient Toxicity Prediction when Annotators Disagree
• Authors: Harbani Jaggi, Kashyap Murali, Eve Fleisig, Erdem Bıyık,
• Abstract summary: When annotators disagree, predicting the labels given by individual annotators can capture nuances overlooked by traditional label aggregation. We introduce three approaches to predicting individual annotator ratings on the toxicity of text. We study the utility of demographic information for rating prediction.
• Score: 1.3749490831384268
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: When annotators disagree, predicting the labels given by individual annotators can capture nuances overlooked by traditional label aggregation. We introduce three approaches to predicting individual annotator ratings on the toxicity of text by incorporating individual annotator-specific information: a neural collaborative filtering (NCF) approach, an in-context learning (ICL) approach, and an intermediate embedding-based architecture. We also study the utility of demographic information for rating prediction. NCF showed limited utility; however, integrating annotator history, demographics, and survey information permits both the embedding-based architecture and ICL to substantially improve prediction accuracy, with the embedding-based architecture outperforming the other methods. We also find that, if demographics are predicted from survey information, using these imputed demographics as features performs comparably to using true demographic data. This suggests that demographics may not provide substantial information for modeling ratings beyond what is captured in survey responses. Our findings raise considerations about the relative utility of different types of annotator information and provide new approaches for modeling annotators in subjective NLP tasks.

Related papers

• Evaluating Facial Expression Recognition Datasets for Deep Learning: A Benchmark Study with Novel Similarity Metrics [4.137346786534721]
  This study investigates the key characteristics and suitability of widely used Facial Expression Recognition (FER) datasets for training deep learning models. We compiled and analyzed 24 FER datasets, including those targeting specific age groups such as children, adults, and the elderly. Benchmark experiments using state-of-the-art neural networks reveal that large-scale, automatically collected datasets tend to generalize better.
  arXiv  Detail & Related papers  (2025-03-26T11:01:00Z)
• Value Profiles for Encoding Human Variation [17.23399556310694]
  We propose representing individuals using value profiles -- natural language descriptions of underlying values compressed from in-context demonstrations. We find that demonstrations contain the most information, followed by value profiles and then demographics. Value profiles offer advantages in terms of scrutability, interpretability, and steerability due to their compressed natural language format.
  arXiv  Detail & Related papers  (2025-03-19T17:57:49Z)
• Fairness without Sensitive Attributes via Knowledge Sharing [13.141672574114597]
  We propose a confidence-based hierarchical classifier structure called "Reckoner" for reliable fair model learning under the assumption of missing sensitive attributes. Our experimental results show that Reckoner consistently outperforms state-of-the-art baselines in COMPAS dataset and New Adult dataset.
  arXiv  Detail & Related papers  (2024-09-27T06:16:14Z)
• Measuring and Improving Attentiveness to Partial Inputs with Counterfactuals [91.59906995214209]
  We propose a new evaluation method, Counterfactual Attentiveness Test (CAT) CAT uses counterfactuals by replacing part of the input with its counterpart from a different example, expecting an attentive model to change its prediction. We show that GPT3 becomes less attentive with an increased number of demonstrations, while its accuracy on the test data improves.
  arXiv  Detail & Related papers  (2023-11-16T06:27:35Z)
• Bring Your Own Data! Self-Supervised Evaluation for Large Language Models [52.15056231665816]
  We propose a framework for self-supervised evaluation of Large Language Models (LLMs) We demonstrate self-supervised evaluation strategies for measuring closed-book knowledge, toxicity, and long-range context dependence. We find strong correlations between self-supervised and human-supervised evaluations.
  arXiv  Detail & Related papers  (2023-06-23T17:59:09Z)
• Metrics for Dataset Demographic Bias: A Case Study on Facial Expression Recognition [4.336779198334903]
  One of the most prominent types of demographic bias are statistical imbalances in the representation of demographic groups in the datasets. We develop a taxonomy for the classification of these metrics, providing a practical guide for the selection of appropriate metrics. The paper provides valuable insights for researchers in AI and related fields to mitigate dataset bias and improve the fairness and accuracy of AI models.
  arXiv  Detail & Related papers  (2023-03-28T11:04:18Z)
• Towards Open-World Feature Extrapolation: An Inductive Graph Learning Approach [80.8446673089281]
  We propose a new learning paradigm with graph representation and learning. Our framework contains two modules: 1) a backbone network (e.g., feedforward neural nets) as a lower model takes features as input and outputs predicted labels; 2) a graph neural network as an upper model learns to extrapolate embeddings for new features via message passing over a feature-data graph built from observed data.
  arXiv  Detail & Related papers  (2021-10-09T09:02:45Z)
• Adversarial Scrubbing of Demographic Information for Text Classification [29.676274451459896]
  We present an adversarial learning framework "Adversarial Scrubber" (ADS), to debias contextual representations. We show that our framework converges without leaking demographic information under certain conditions.
  arXiv  Detail & Related papers  (2021-09-17T15:38:43Z)
• Predicting Census Survey Response Rates With Parsimonious Additive Models and Structured Interactions [14.003044924094597]
  We consider the problem of predicting survey response rates using a family of flexible and interpretable nonparametric models. The study is motivated by the US Census Bureau's well-known ROAM application.
  arXiv  Detail & Related papers  (2021-08-24T17:49:55Z)
• Learning Bias-Invariant Representation by Cross-Sample Mutual Information Minimization [77.8735802150511]
  We propose a cross-sample adversarial debiasing (CSAD) method to remove the bias information misused by the target task. The correlation measurement plays a critical role in adversarial debiasing and is conducted by a cross-sample neural mutual information estimator. We conduct thorough experiments on publicly available datasets to validate the advantages of the proposed method over state-of-the-art approaches.
  arXiv  Detail & Related papers  (2021-08-11T21:17:02Z)
• Balancing Biases and Preserving Privacy on Balanced Faces in the Wild [50.915684171879036]
  There are demographic biases present in current facial recognition (FR) models. We introduce our Balanced Faces in the Wild dataset to measure these biases across different ethnic and gender subgroups. We find that relying on a single score threshold to differentiate between genuine and imposters sample pairs leads to suboptimal results. We propose a novel domain adaptation learning scheme that uses facial features extracted from state-of-the-art neural networks.
  arXiv  Detail & Related papers  (2021-03-16T15:05:49Z)
• A Revised Generative Evaluation of Visual Dialogue [80.17353102854405]
  We propose a revised evaluation scheme for the VisDial dataset. We measure consensus between answers generated by the model and a set of relevant answers. We release these sets and code for the revised evaluation scheme as DenseVisDial.
  arXiv  Detail & Related papers  (2020-04-20T13:26:45Z)
• Learning Unbiased Representations via Mutual Information Backpropagation [36.383338079229695]
  In particular, we face the case where some attributes (bias) of the data, if learned by the model, can severely compromise its generalization properties. We propose a novel end-to-end optimization strategy, which simultaneously estimates and minimizes the mutual information between the learned representation and the data attributes.
  arXiv  Detail & Related papers  (2020-03-13T18:06:31Z)

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.