# What Triggers my Model? Contrastive Explanations Inform Gender Choices by Translation Models

arXiv: https://arxiv.org/abs/2512.08440 · HTML: https://mlsanity.com/paper/2512.08440/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 35 of its 61 references that are in the map.
- First posted on arXiv 9 Dec 2025.

## Builds on (most cited first, 30 of 35)

- [Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings](https://mlsanity.com/paper/1607.06520/) (2016)
- [Language (Technology) is Power: A Critical Survey of "Bias" in NLP](https://mlsanity.com/paper/2005.14050/) (2020)
- [Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness?](https://mlsanity.com/paper/2004.03685/) (2020)
- [Evaluating Gender Bias in Machine Translation](https://mlsanity.com/paper/1906.00591/) (2019)
- [Gender Bias in Contextualized Word Embeddings](https://mlsanity.com/paper/1904.03310/) (2019)
- [Explainable Machine Learning for Scientific Insights and Discoveries](https://mlsanity.com/paper/1905.08883/) (2019)
- [Post-hoc Interpretability for Neural NLP: A Survey](https://mlsanity.com/paper/2108.04840/) (2021)
- [The elephant in the interpretability room: Why use attention as explanation when we have saliency methods?](https://mlsanity.com/paper/2010.05607/) (2020)
- [Context-Aware Neural Machine Translation Learns Anaphora Resolution](https://mlsanity.com/paper/1805.10163/) (2018)
- [Evaluating the Underlying Gender Bias in Contextualized Word Embeddings](https://mlsanity.com/paper/1904.08783/) (2019)
- [Interpreting Language Models with Contrastive Explanations](https://mlsanity.com/paper/2202.10419/) (2022)
- [Reducing Gender Bias in Neural Machine Translation as a Domain Adaptation Problem](https://mlsanity.com/paper/2004.04498/) (2020)
- [Inseq: An Interpretability Toolkit for Sequence Generation Models](https://mlsanity.com/paper/2302.13942/) (2023)
- [An Evaluation of the Human-Interpretability of Explanation](https://mlsanity.com/paper/1902.00006/) (2019)
- ["Will You Find These Shortcuts?" A Protocol for Evaluating the Faithfulness of Input Salience Methods for Text Classification](https://mlsanity.com/paper/2111.07367/) (2021)
- [Collecting a Large-Scale Gender Bias Dataset for Coreference Resolution and Machine Translation](https://mlsanity.com/paper/2109.03858/) (2021)
- [On Measuring Gender Bias in Translation of Gender-neutral Pronouns](https://mlsanity.com/paper/1905.11684/) (2019)
- [Gender in Danger? Evaluating Speech Translation Technology on the MuST-SHE Corpus](https://mlsanity.com/paper/2006.05754/) (2020)
- [NeuTral Rewriter: A Rule-Based and Neural Approach to Automatic Rewriting into Gender-Neutral Alternatives](https://mlsanity.com/paper/2109.06105/) (2021)
- [How Grammatical is Character-level Neural Machine Translation? Assessing MT Quality with Contrastive Translation Pairs](https://mlsanity.com/paper/1612.04629/) (2016)
- [Multilingual Holistic Bias: Extending Descriptors and Patterns to Unveil Demographic Biases in Languages at Scale](https://mlsanity.com/paper/2305.13198/) (2023)
- [Gender Neutralization for an Inclusive Machine Translation: from Theoretical Foundations to Open Challenges](https://mlsanity.com/paper/2301.10075/) (2023)
- [Gender Coreference and Bias Evaluation at WMT 2020](https://mlsanity.com/paper/2010.06018/) (2020)
- [Do Context-Aware Translation Models Pay the Right Attention?](https://mlsanity.com/paper/2105.06977/) (2021)
- [A Tale of Pronouns: Interpretability Informs Gender Bias Mitigation for Fairer Instruction-Tuned Machine Translation](https://mlsanity.com/paper/2310.12127/) (2023)
- [GENder-IT: An Annotated English-Italian Parallel Challenge Set for Cross-Linguistic Natural Gender Phenomena](https://mlsanity.com/paper/2108.02854/) (2021)
- [Exploiting Biased Models to De-bias Text: A Gender-Fair Rewriting Model](https://mlsanity.com/paper/2305.11140/) (2023)
- [A Prompt Response to the Demand for Automatic Gender-Neutral Translation](https://mlsanity.com/paper/2402.06041/) (2024)
- [Mind the Inclusivity Gap: Multilingual Gender-Neutral Translation Evaluation with mGeNTE](https://mlsanity.com/paper/2501.09409/) (2025)
- [Gender, names and other mysteries: Towards the ambiguous for gender-inclusive translation](https://mlsanity.com/paper/2306.04573/) (2023)

## Built on by (most cited first, 2 of 2)

- [Neuron-Level Interventions for Gendered and Gender-Neutral Generation in Language Models](https://mlsanity.com/paper/2605.30717/) (2026)
- [Explaining GAND: A Resource on Gender-Ambiguous Natural Data & Contrastive Attribution](https://mlsanity.com/paper/2607.22546/) (2026)

## Abstract (arXiv)

Interpretability can be implemented to understand decisions taken by (black box) models, such as neural machine translation (NMT) or large language models (LLMs). Yet, research in this area has been limited in relation to a manifested problem in these models: gender bias. In this work, we aim to move away from simply measuring bias to exploring its origins. Working with gender-ambiguous natural source data, this exploratory study examines which context, in the form of input tokens in the source sentence (EN), influences (or triggers) the NMT model's choice of a certain gender inflection in the target languages (DE/ES). To analyse this, we compute saliency attribution based on contrastive translations. We first address the challenge of the lack of a scoring threshold and specifically examine different attribution levels of source words on the model's gender decisions in the translation. We compare salient source words with human perceptions of gender and demonstrate a noticeable overlap between human perceptions and model attribution. Additionally, we provide a linguistic analysis of salient words. Our work showcases the relevance of understanding model translation decisions in terms of gender, how this compares to human decisions and that this information should be leveraged to mitigate gender bias.