An Analysis of the Effects of Decoding Algorithms on Fairness in Open-Ended Language Generation
Several prior works have shown that language models (LMs) can generate text containing harmful social biases and stereotypes. While decoding algorithms play a central role in determining properties of LM generated text, their impact on the fairness of the generations has not been studied. We present a systematic analysis of the impact of decoding algorithms on LM fairness, and analyze the trade-off between fairness, diversity and quality. Our experiments with top-$p$, top-$k$ and temperature decoding algorithms, in open-ended language generation, show that fairness across demographic groups changes significantly with change in decoding algorithm's hyper-parameters. Notably, decoding algorithms that output more diverse text also output more texts with negative sentiment and regard. We present several findings and provide recommendations on standardized reporting of decoding details in fairness evaluations and optimization of decoding algorithms for fairness alongside quality and diversity.
Code (0)
등록된 구현이 없습니다.
Tasks
DiversityFairnessText GenerationSimilar Papers 제목 키워드 기반
Algorithms for Fairness in Sequential Decision Making
It has recently been shown that if feedback effects of decisions are ignored, then imposing fairness constraints such as demographic parity or equality of opportunity can actually exacerbate unfairness. We propose to add…
Decision MakingFairnessReinforcement LearningSequential Decision MakingEvaluating Causal Discovery Algorithms for Path-Specific Fairness and Utility in Healthcare
Causal discovery in health data faces evaluation challenges when ground truth is unknown. We address this by collaborating with experts to construct proxy ground-truth graphs, establishing benchmarks for synthetic Alzhei…
Causal InferenceCounterfactual Methods for Detecting Unfairness in Anti-Money Laundering Algorithms
The application of machine learning-based predictive algorithms to Anti-Money Laundering (AML) has grown rapidly, driven by the vast volume of financial transaction data available to banks. These algorithms are typically…
Societal Biases in Language Generation: Progress and Challenges
Technology for language generation has advanced rapidly, spurred by advancements in pre-training large models on massive amounts of data and the need for intelligent agents to communicate in a natural manner. While techn…
FairnessText GenerationThe Disparate Impacts of Speculative Decoding
The practice of speculative decoding, whereby inference is probabilistically supported by a smaller, cheaper, ``drafter'' model, has become a standard technique for systematically reducing the decoding time of large lang…