Automate or Assist? The Role of Computational Models in Identifying Gendered Discourse in US Capital Trial Transcripts
- URL: http://arxiv.org/abs/2407.12500v2
- Date: Sat, 27 Jul 2024 00:52:44 GMT
- Title: Automate or Assist? The Role of Computational Models in Identifying Gendered Discourse in US Capital Trial Transcripts
- Authors: Andrea W Wen-Yi, Kathryn Adamson, Nathalie Greenfield, Rachel Goldberg, Sandra Babcock, David Mimno, Allison Koenecke,
- Abstract summary: We present a case study of adding a computational model to a complex problem: identifying gender-biased language in US capital trials for women defendants.
Unlike many typical NLP tasks, annotating for gender bias in months-long capital trials is complicated, with many individual judgment calls.
This experience suggests that seeking to replace experts with computational models for complex annotation is both unrealistic and undesirable.
- Score: 2.790550055247773
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: The language used by US courtroom actors in criminal trials has long been studied for biases. However, systematic studies for bias in high-stakes court trials have been difficult, due to the nuanced nature of bias and the legal expertise required. Large language models offer the possibility to automate annotation. But validating the computational approach requires both an understanding of how automated methods fit in existing annotation workflows and what they really offer. We present a case study of adding a computational model to a complex and high-stakes problem: identifying gender-biased language in US capital trials for women defendants. Our team of experienced death-penalty lawyers and NLP technologists pursue a three-phase study: first annotating manually, then training and evaluating computational models, and finally comparing expert annotations to model predictions. Unlike many typical NLP tasks, annotating for gender bias in months-long capital trials is complicated, with many individual judgment calls. Contrary to standard arguments for automation that are based on efficiency and scalability, legal experts find the computational models most useful in providing opportunities to reflect on their own bias in annotation and to build consensus on annotation rules. This experience suggests that seeking to replace experts with computational models for complex annotation is both unrealistic and undesirable. Rather, computational models offer valuable opportunities to assist the legal experts in annotation-based studies.
Related papers
- A linguistically-motivated evaluation methodology for unraveling model's abilities in reading comprehension tasks [10.181408678232055]
We introduce an evaluation methodology for reading comprehension tasks based on the intuition that certain examples consistently yield lower scores regardless of model size or architecture.
We capitalize on semantic frame annotation for characterizing this complexity, and study seven complexity factors that may account for model's difficulty.
arXiv Detail & Related papers (2025-01-29T11:05:20Z) - Foundation Models at Work: Fine-Tuning for Fairness in Algorithmic Hiring [5.482898079941062]
We present AutoRefine, a method that leverages reinforcement learning for targeted fine-tuning.
We demonstrate the method for a problem arising in algorithmic hiring platforms where linguistic biases influence a recommendation system.
Our model detects and regulates biases in job descriptions to meet diversity and fairness criteria.
arXiv Detail & Related papers (2025-01-13T13:36:17Z) - Whither Bias Goes, I Will Go: An Integrative, Systematic Review of Algorithmic Bias Mitigation [1.0470286407954037]
Concerns have been raised that machine learning (ML) models may be biased and perpetuate or exacerbate inequality.
We present a four-stage model of developing ML assessments and applying bias mitigation methods.
arXiv Detail & Related papers (2024-10-21T02:32:14Z) - Revisiting the Superficial Alignment Hypothesis [0.9831489366502302]
The Superficial Alignment Hypothesis posits that almost all of a language model's abilities and knowledge are learned during pre-training.
We re-examine these claims by studying the scaling behavior of post-training with increasing finetuning examples.
arXiv Detail & Related papers (2024-09-27T22:14:10Z) - Lawma: The Power of Specialization for Legal Annotation [18.45967769381101]
We introduce CaselawQA, a benchmark comprising 260 legal annotation tasks.
We demonstrate that commercial models, such as GPT-4.5 and Claude 3.7 Sonnet, achieve non-trivial yet highly variable accuracy.
We then demonstrate that small, lightly fine-tuned models outperform commercial models.
arXiv Detail & Related papers (2024-07-23T16:23:04Z) - The Art of Saying No: Contextual Noncompliance in Language Models [123.383993700586]
We introduce a comprehensive taxonomy of contextual noncompliance describing when and how models should not comply with user requests.
Our taxonomy spans a wide range of categories including incomplete, unsupported, indeterminate, and humanizing requests.
To test noncompliance capabilities of language models, we use this taxonomy to develop a new evaluation suite of 1000 noncompliance prompts.
arXiv Detail & Related papers (2024-07-02T07:12:51Z) - Auditing the Use of Language Models to Guide Hiring Decisions [2.949890760187898]
Regulatory efforts to protect against algorithmic bias have taken on increased urgency with rapid advances in large language models.
Current regulations -- as well as the scientific literature -- provide little guidance on how to conduct these assessments.
Here we propose and investigate one approach for auditing algorithms: correspondence experiments.
arXiv Detail & Related papers (2024-04-03T22:01:26Z) - Self-Debiasing Large Language Models: Zero-Shot Recognition and
Reduction of Stereotypes [73.12947922129261]
We leverage the zero-shot capabilities of large language models to reduce stereotyping.
We show that self-debiasing can significantly reduce the degree of stereotyping across nine different social groups.
We hope this work opens inquiry into other zero-shot techniques for bias mitigation.
arXiv Detail & Related papers (2024-02-03T01:40:11Z) - Modeling Legal Reasoning: LM Annotation at the Edge of Human Agreement [3.537369004801589]
We study the classification of legal reasoning according to jurisprudential philosophy.
We use a novel dataset of historical United States Supreme Court opinions annotated by a team of domain experts.
We find that generative models perform poorly when given instructions equal to the instructions presented to human annotators.
arXiv Detail & Related papers (2023-10-27T19:27:59Z) - Evaluating Language Models for Mathematics through Interactions [116.67206980096513]
We introduce CheckMate, a prototype platform for humans to interact with and evaluate large language models (LLMs)
We conduct a study with CheckMate to evaluate three language models (InstructGPT, ChatGPT, and GPT-4) as assistants in proving undergraduate-level mathematics.
We derive a taxonomy of human behaviours and uncover that despite a generally positive correlation, there are notable instances of divergence between correctness and perceived helpfulness.
arXiv Detail & Related papers (2023-06-02T17:12:25Z) - 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) - Fairness-guided Few-shot Prompting for Large Language Models [93.05624064699965]
In-context learning can suffer from high instability due to variations in training examples, example order, and prompt formats.
We introduce a metric to evaluate the predictive bias of a fixed prompt against labels or a given attributes.
We propose a novel search strategy based on the greedy search to identify the near-optimal prompt for improving the performance of in-context learning.
arXiv Detail & Related papers (2023-03-23T12:28:25Z) - Investigating Fairness Disparities in Peer Review: A Language Model
Enhanced Approach [77.61131357420201]
We conduct a thorough and rigorous study on fairness disparities in peer review with the help of large language models (LMs)
We collect, assemble, and maintain a comprehensive relational database for the International Conference on Learning Representations (ICLR) conference from 2017 to date.
We postulate and study fairness disparities on multiple protective attributes of interest, including author gender, geography, author, and institutional prestige.
arXiv Detail & Related papers (2022-11-07T16:19:42Z) - On the Basis of Sex: A Review of Gender Bias in Machine Learning
Applications [0.0]
We first introduce several examples of machine learning gender bias in practice.
We then detail the most widely used formalizations of fairness in order to address how to make machine learning models fairer.
arXiv Detail & Related papers (2021-04-06T14:11:16Z)
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.