paper-with-me

홈 › Papers

Real Negatives Matter: Continuous Training with Real Negatives for Delayed Feedback Modeling

2021-04-29 · Siyu Gu, Xiang-Rong Sheng, Ying Fan, Guorui Zhou, Xiaoqiang Zhu

One of the difficulties of conversion rate (CVR) prediction is that the conversions can delay and take place long after the clicks. The delayed feedback poses a challenge: fresh data are beneficial to continuous training but may not have complete label information at the time they are ingested into the training pipeline. To balance model freshness and label certainty, previous methods set a short waiting window or even do not wait for the conversion signal. If conversion happens outside the waiting window, this sample will be duplicated and ingested into the training pipeline with a positive label. However, these methods have some issues. First, they assume the observed feature distribution remains the same as the actual distribution. But this assumption does not hold due to the ingestion of duplicated samples. Second, the certainty of the conversion action only comes from the positives. But the positives are scarce as conversions are sparse in commercial systems. These issues induce bias during the modeling of delayed feedback. In this paper, we propose DElayed FEedback modeling with Real negatives (DEFER) method to address these issues. The proposed method ingests real negative samples into the training pipeline. The ingestion of real negatives ensures the observed feature distribution is equivalent to the actual distribution, thus reducing the bias. The ingestion of real negatives also brings more certainty information of the conversion. To correct the distribution shift, DEFER employs importance sampling to weigh the loss function. Experimental results on industrial datasets validate the superiority of DEFER. DEFER have been deployed in the display advertising system of Alibaba, obtaining over 6.0% improvement on CVR in several scenarios. The code and data in this paper are now open-sourced {https://github.com/gusuperstar/defer.git}.

📄 PDF Abstract BibTeX arXiv:2104.14121

Code (1)

gusuperstar/defer 공식 구현 tf

Similar Papers 제목 키워드 기반

Real or Fake? Learning to Discriminate Machine from Human Generated Text

2019-06-07 · Anton Bakhtin, Sam Gross, Myle Ott, Yuntian Deng 외

Energy-based models (EBMs), a.k.a. un-normalized models, have had recent successes in continuous spaces. However, they have not been successfully applied to model text sequences. While decreasing the energy at training s…

Language ModelingLanguage ModellingText Generation

Selective Contrastive Learning For Gloss Free Sign Language Translation

2026-04-24 · Changhao Lai, Rui Zhao, Xuewen Zhong, Jinsong Su 외 arxiv

Sign language translation (SLT) converts continuous sign videos into spoken-language text, yet it remains challenging due to the intrinsic modality mismatch between visual signs and written text, particularly in gloss-fr…

Sign Language TranslationContrastive Learning

Vision Language Pre-training by Contrastive Learning with Cross-Modal Similarity Regulation

2023-05-08 · Chaoya Jiang, Wei Ye, Haiyang Xu, Miang yan 외

Cross-modal contrastive learning in vision language pretraining (VLP) faces the challenge of (partial) false negatives. In this paper, we study this problem from the perspective of Mutual Information (MI) optimization. I…

Common Sense ReasoningContrastive Learning

Exploiting Pseudo Image Captions for Multimodal Summarization

2023-05-09 · Chaoya Jiang, Rui Xie, Wei Ye, Jinan Sun 외

Cross-modal contrastive learning in vision language pretraining (VLP) faces the challenge of (partial) false negatives. In this paper, we study this problem from the perspective of Mutual Information (MI) optimization. I…

Common Sense ReasoningContrastive LearningImage Captioning

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 외 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 c…

Knowledge DistillationContrastive Learning