paper-with-me

홈 › Papers

WebFAQ 2.0: A Multilingual QA Dataset with Mined Hard Negatives for Dense Retrieval

2026-02-19 · Michael Dinzinger, Laura Caspari, Ali Salman, Irvin Topi, Jelena Mitrović, Michael Granitzer arxiv

We introduce WebFAQ 2.0, a new version of the WebFAQ dataset, containing 198 million FAQ-based natural question-answer pairs across 108 languages. Compared to the previous version, it significantly expands multilingual coverage and the number of bilingual aligned QA pairs to over 14.3M, making it the largest FAQ-based resource. Unlike the original release, WebFAQ 2.0 uses a novel data collection strategy that directly crawls and extracts relevant web content, resulting in a substantially more diverse and multilingual dataset with richer context through page titles and descriptions. In response to community feedback, we also release a hard negatives dataset for training dense retrievers, with 1.25M queries across 20 languages. These hard negatives were mined using a two-stage retrieval pipeline and include cross-encoder scores for 200 negatives per query. We further show how this resource enables two primary fine-tuning strategies for dense retrievers: Contrastive Learning with MultipleNegativesRanking loss, and Knowledge Distillation with MarginMSE loss. WebFAQ 2.0 is not a static resource but part of a long-term effort. Since late 2025, structured FAQs are being regularly released through the Open Web Index, enabling continuous expansion and refinement. We publish the datasets and training scripts to facilitate further research in multilingual and cross-lingual IR. The dataset itself and all related resources are publicly available on GitHub and HuggingFace.

📄 PDF Abstract BibTeX arXiv:2602.17327

Code (0)

등록된 구현이 없습니다.

Tasks

Knowledge DistillationContrastive Learning

Similar Papers 제목 키워드 기반

WebFAQ: A Multilingual Collection of Natural Q&A Datasets for Dense Retrieval

2025-02-28 · Michael Dinzinger, Laura Caspari, Kanishka Ghosh Dastidar, Jelena Mitrović 외

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) pair…

Dataset GenerationOpen-Domain Question AnsweringQuestion AnsweringRetrieval+1

Remining Hard Negatives for Generative Pseudo Labeled Domain Adaptation

2025-01-24 · Goksenin Yuksel, David Rau, Jaap Kamps

Dense retrievers have demonstrated significant potential for neural information retrieval; however, they exhibit a lack of robustness to domain shifts, thereby limiting their efficacy in zero-shot settings across diverse…

Domain AdaptationInformation RetrievalKnowledge Distillation

Adaptive Offline Quintuplet Loss for Image-Text Matching

2020-03-07 · ECCV 2020 8 · Tianlang Chen, Jiajun Deng, Jiebo Luo

Existing image-text matching approaches typically leverage triplet loss with online hard negatives to train the model. For each image or text anchor in a training mini-batch, the model is trained to distinguish between a…

Image-text matchingText MatchingTriplet

Improving Visual-Semantic Embeddings by Learning Semantically-Enhanced Hard Negatives for Cross-modal Information Retrieval

2022-10-10 · Yan Gong, Georgina Cosma

Visual Semantic Embedding (VSE) aims to extract the semantics of images and their descriptions, and embed them into the same latent space for cross-modal information retrieval. Most existing VSE networks are trained by a…

Cross-Modal Information RetrievalImage DescriptionInformation RetrievalRetrieval

Global Contrastive Batch Sampling via Optimization on Sample Permutations

2022-10-23 · Vin Sachidananda, ZiYi Yang, Chenguang Zhu

Contrastive Learning has recently achieved state-of-the-art performance in a wide range of tasks. Many contrastive learning approaches use mined hard negatives to make batches more informative during training but these a…

Code SearchContrastive LearningSentenceSentence Embedding+1