Quantifying Bias from Decoding Techniques in Natural Language Generation

Mayukh Das, Wolf Tilo Balke


Abstract
Natural language generation (NLG) models can propagate social bias towards particular demography. Though several studies investigated bias from data and model, NLG task distinctively uses stochastic decoder that can positively or negatively impact the bias-sensitive tokens initially predicted by the model. To address this gap in research, we present an extensive analysis of bias from decoding techniques for open-domain language generation considering the entire decoding space. We analyze to what extent bias metrics like toxicity and sentiment are impacted by the individual components of decoder algorithms. To this extent, we also analyze the trade-off between bias scores and human-annotated generation quality throughout the decoder space. Together, these methods reveal the imperative of testing inference time bias and provide evidence on the usefulness of inspecting the entire decoding spectrum.
Anthology ID:
2022.coling-1.112
Volume:
Proceedings of the 29th International Conference on Computational Linguistics
Month:
October
Year:
2022
Address:
Gyeongju, Republic of Korea
Venue:
COLING
SIG:
Publisher:
International Committee on Computational Linguistics
Note:
Pages:
1311–1323
Language:
URL:
https://aclanthology.org/2022.coling-1.112
DOI:
Bibkey:
Cite (ACL):
Mayukh Das and Wolf Tilo Balke. 2022. Quantifying Bias from Decoding Techniques in Natural Language Generation. In Proceedings of the 29th International Conference on Computational Linguistics, pages 1311–1323, Gyeongju, Republic of Korea. International Committee on Computational Linguistics.
Cite (Informal):
Quantifying Bias from Decoding Techniques in Natural Language Generation (Das & Balke, COLING 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.coling-1.112.pdf
Code
 mayukhga83/decoder-bias
Data
GAP Coreference DatasetWebText