Logo Lanfrica
  • Home
  • Atlas
  • Insights
  • Docs
  • Sign in

© 2026 Lanfrica. All rights reserved. All copyrights of the resources shown on the Lanfrica website belong to the original copyright holders, unless explicitly stated otherwise.

Cross-Lingual Robustness of WebFAQ Retrieval Models Against Paraphrase Attacks in Low-Resource Language Families

Domain:

natural language processing

Record type:

dataset
Creator:
SOV
Publisher:
Zenodo
Host:avatar
We present WebFAQ, a large-scale collection of open-domain question answering datasets derived from FAQ-style schema.org annotations. In total, the data collection consists of 96 million natural question-answer (QA) pairs across 75 languages, including 47 million (49\%) non-English samples. WebFAQ further serves as the foundation for 20 monolingual retrieval benchmarks with a total size of 11.2 million QA pairs (5.9 million non-English). These datasets are carefully curated through refined filtering and near-duplicate detection, yielding high-quality resources for training and evaluating multil Research goal: How does the performance of multilingual retrieval models trained on WebFAQ datasets vary across different language families when evaluated using paraphrase-based adversarial attacks, and what metrics (e.g., nDCG, MRR) show the largest degradation for low-resource languages? Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 7.6/10. This report was generated autonomously by SOVEREIGN Research Kernel, an owner-gated autonomous research lab. The content synthesizes findings from peer-reviewed papers. Tribunal score: 7.6/10.

Visit

doi.orgzenodo.org

Tasks

information retrievalquestion answering

Tags

performancemultilingualretrievalmodelstrainedWebFAQdatasetsvary

Licenses

Creative Commons Attribution 4.0 Internationalhttps://creativecommons.org/licenses/by/4.0/legalcode

Similar

Robustness of Zero-Shot Cross-Lingual Retrieval Models Against Adversarial Noise in Low-Resource Language PairsDegradation of Cross-Lingual Dense Retrieval MRR on Low-Resource WebFAQ Language Families Versus Monolingual BaselinesRobustness of Zero-Shot Cross-Lingual Retrieval Models Against Domain Shift in Low-Resource Languages via ArtificialHybrid Batch Training for Cross-Lingual Retrieval Robustness in Low-Resource Multilingual Pretrained ModelsRobustness of Zero-Shot Cross-Lingual Retrieval Models Trained on Code-Switched Data Across Language FamiliesCross-lingual Zero-shot Retrieval Accuracy of WebFAQ vs. TyDi QA Models on Low-resource XQuAD Subsets

Robustness of Zero-Shot Cross-Lingual Retrieval Models Against Adversarial Noise in Low-Resource Language Pairs

Transferring information retrieval (IR) models from a high-resource language (typically English) to

Degradation of Cross-Lingual Dense Retrieval MRR on Low-Resource WebFAQ Language Families Versus Monolingual Baselines

We present WebFAQ, a large-scale collection of open-domain question answering datasets derived from

Robustness of Zero-Shot Cross-Lingual Retrieval Models Against Domain Shift in Low-Resource Languages via Artificial

Transferring information retrieval (IR) models from a high-resource language (typically English) to

Hybrid Batch Training for Cross-Lingual Retrieval Robustness in Low-Resource Multilingual Pretrained Models

Benefiting from transformer-based pre-trained language models, neural ranking models have made signi

Robustness of Zero-Shot Cross-Lingual Retrieval Models Trained on Code-Switched Data Across Language Families

Transferring information retrieval (IR) models from a high-resource language (typically English) to

Cross-lingual Zero-shot Retrieval Accuracy of WebFAQ vs. TyDi QA Models on Low-resource XQuAD Subsets

Benefiting from transformer-based pre-trained language models, neural ranking models have made signi