paper-with-me

홈 › Papers

Block-SCL: Blocking Matters for Supervised Contrastive Learning in Product Matching

2022-07-05 · Mario Almagro, David Jiménez, Diego Ortego, Emilio Almazán, Eva Martínez

Product matching is a fundamental step for the global understanding of consumer behavior in e-commerce. In practice, product matching refers to the task of deciding if two product offers from different data sources (e.g. retailers) represent the same product. Standard pipelines use a previous stage called blocking, where for a given product offer a set of potential matching candidates are retrieved based on similar characteristics (e.g. same brand, category, flavor, etc.). From these similar product candidates, those that are not a match can be considered hard negatives. We present Block-SCL, a strategy that uses the blocking output to make the most of Supervised Contrastive Learning (SCL). Concretely, Block-SCL builds enriched batches using the hard-negatives samples obtained in the blocking stage. These batches provide a strong training signal leading the model to learn more meaningful sentence embeddings for product matching. Experimental results in several public datasets demonstrate that Block-SCL achieves state-of-the-art results despite only using short product titles as input, no data augmentation, and a lighter transformer backbone than competing methods.

📄 PDF Abstract BibTeX arXiv:2207.02008

Code (0)

등록된 구현이 없습니다.

Tasks

BlockingContrastive LearningData AugmentationSentenceSentence Embeddings

Methods 이 논문이 사용한 방법론

Contrastive Learning 설명 없음

Similar Papers 제목 키워드 기반

Towards Universal Dense Blocking for Entity Resolution

2024-04-23 · Tianshu Wang, Hongyu Lin, Xianpei Han, Xiaoyang Chen 외

Blocking is a critical step in entity resolution, and the emergence of neural network-based representation models has led to the development of dense blocking as a promising approach for exploring deep semantics in block…

BlockingContrastive LearningEntity Resolution

SC-Block: Supervised Contrastive Blocking within Entity Resolution Pipelines

2023-03-06 · Alexander Brinkmann, Roee Shraga, Christian Bizer

The goal of entity resolution is to identify records in multiple datasets that represent the same real-world entity. However, comparing all records across datasets can be computationally intensive, leading to long runtim…

BlockingContrastive LearningEntity Resolution

ShallowBlocker: Improving Set Similarity Joins for Blocking

2023-12-26 · Nils Barlaug

Blocking is a crucial step in large-scale entity matching but often requires significant manual engineering from an expert for each new dataset. Recent work has show that deep learning is state-of-the-art and has great p…

BlockingDeep Learning

Sparkly: A Simple yet Surprisingly Strong TF/IDF Blocker for Entity Matching

2023-04-20 · Proceedings of the VLDB Endowment 2023 4 · Derek Paulsen, Yash Govind, AnHai Doan

Blocking is a major task in entity matching. Numerous blocking solutions have been developed, but as far as we can tell, blocking using the well-known tf/idf measure has received virtually no attention. Yet, when we expe…

Blocking

Multi-Modality Deep Network for JPEG Artifacts Reduction

2023-05-04 · Xuhao Jiang, Weimin Tan, Qing Lin, Chenxi Ma 외

In recent years, many convolutional neural network-based models are designed for JPEG artifacts reduction, and have achieved notable progress. However, few methods are suitable for extreme low-bitrate image compression a…

Contrastive LearningImage CompressionImage Deblocking