Exploration and Mitigation of Stereotypical Gender Biases in Information Retrieval Systems
Bibliographic record
Abstract
Recent studies in information retrieval have shown that gender biases have found their way into representational and algorithmic aspects of retrieval methods. In this thesis, we focus on the exploration and mitigation of gender biases in information retrieval gold standard datasets, often referred to as relevance judgements, and di▯erent retrieval methods including ad hoc retrieval and neural rankers. We investigate the presence of stereotypical gender biases in relevance judgment datasets and show that stereotypical gender biases are prevalent in relevance judgement datasets. The presence of gender biases in relevance judgements would immediately find its way into how neural ranking models are trained and evaluated. Therefore, we propose a systematic method to de-bias the relevance judgement datasets with a set of balanced and well-matched query pairs from different gender identities to reduce the level of bias transferred into neural ranking models that are trained based on them. The main premise of our work is that such a debiasing process will expose neural rankers to comparable queries from across gender identities that have associated relevant documents with compatible degrees of gender bias. Our experiments show that our proposed approach is able to systematically reduce stereotypical gender biases associated with different gender identities, and at the same time maintain the same level of retrieval effectiveness. In addition to relevance judgment datasets, we propose methods for reducing biases in different retrieval methods. To this end, we propose a bias-aware pseudo-relevance feedback method for reducing the level of bias in the ad hoc retrieval method and show that this decrease in bias is not accompanied with the cost of reduction in the utility. We also propose a light-weight strategy that considers the degrees of gender bias when sampling documents to be used for training neural ranking models. Our results show that the proposed light-weight strategy is able to show competitive (or even better) performance compared to the state-of-the-art neural architectures specifically designed to reduce gender biases.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.011 | 0.044 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".