Artificial Intelligence 8 min read

Alibaba's Five Papers Accepted at SIGIR 2022

Alibaba’s research team had five papers accepted at the prestigious SIGIR 2022 conference in Madrid, covering innovations such as joint ad‑ranking and creative selection, personalized bundle generation, calibrated neural predictions, disentangled counterfactual regression, and cold‑start user recommendation, showcasing strong expertise in information retrieval and online advertising.

Alimama Tech
Alimama Tech
Alimama Tech
Alibaba's Five Papers Accepted at SIGIR 2022

Alibaba's technical team has achieved significant success at the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2022), with five papers accepted for presentation. SIGIR 2022, a top-tier conference recommended by the China Computer Federation (CCF), will be held in Madrid, Spain from July 11-15, 2022, with both in-person and virtual attendance options.

The conference received 794 long paper submissions with 161 accepted (20% acceptance rate) and 667 short paper submissions with 165 accepted (24.7% acceptance rate). The five accepted papers from Alibaba's team cover various cutting-edge topics in information retrieval and online advertising.

The first paper, "Joint Optimization of Ad Ranking and Creative Selection," proposes a novel cascade architecture for creative selection that integrates ad ranking and creative optimization. The second paper, "Towards Personalized Bundle Creative Generation with Contrastive Non-Autoregressive Decoding," introduces a model for generating bundled creative content that considers both quality and generation speed. The third paper, "Posterior Probability Matters: Doubly-Adaptive Calibration for Neural Predictions in Online Advertising," presents AdaCalib, a calibration model that improves prediction accuracy in online advertising systems.

The fourth paper, "Learning Disentangled Representations for Counterfactual Regression via Mutual Information Minimization," addresses the challenge of learning disentangled representations for counterfactual inference through mutual information minimization. The fifth paper, "Transform Cold-Start Users into Warm via Fused Behaviors in Large-Scale Recommendation," proposes a recommendation model that effectively handles cold-start users by fusing behavior sequences and adapting embeddings.

These papers demonstrate Alibaba's strong research capabilities in information retrieval, machine learning, and online advertising, contributing valuable insights to the academic community and practical applications in the industry.

Recommendation systemsinformation retrievalmachine learningonline advertisingcalibrationcold-start problemcounterfactual inferencecreative generationSIGIR 2022
Alimama Tech
Written by

Alimama Tech

Official Alimama tech channel, showcasing all of Alimama's technical innovations.

0 followers
Reader feedback

How this landed with the community

login Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.