Bridging LLMs and Display Advertising, Scaling Laws for Online Advertisement Systems, and More!

Vol.79 for Nov 18 - Nov 24, 2024

Stay Ahead of the Curve with the Latest Advancements and Discoveries in Information Retrieval.

This week’s newsletter highlights the following research:

1. Scaling Transformer Architecture for Billion-Scale Short Video Retrieval, from Kuaishou

2. Understanding the Limitations of Cross-Encoder Rerankers, from Databricks

3. Learning Trainable Multi-Modal Features for Large-Scale Recommendation Systems, from Kuaishou

4. Understanding and Applying Scaling Laws in Commercial Advertisement Retrieval System, from Kuaishou

5. Bridging LLMs and Commercial Display Advertising at Scale, from Tencent

6. Writing Style Bias in Text Embedding Models for Information Retrieval, from Amadeus

7. A Hard Negative Enhanced Framework for Contrastive Sentence Embeddings, from Liu et al.

8. Efficient Processing of Network Data through Schema-Driven Retrieval Augmentation, from NIST

9. A Novel Architecture for Heterogeneous CTR Prediction, from Meta

10. Adaptive Query Rewriting for Improved Retrieval-Augmented Generation, from Kuaishou