Saved in:
Bibliographic Details
Main Authors: Rajagopalan, Santosh, Vronsky, Jonathan, Yan, Songbai, Golestaneh, S. Alireza, Chandra, Shubhra, Zhou, Min
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.18785
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917176141676544
author Rajagopalan, Santosh
Vronsky, Jonathan
Yan, Songbai
Golestaneh, S. Alireza
Chandra, Shubhra
Zhou, Min
author_facet Rajagopalan, Santosh
Vronsky, Jonathan
Yan, Songbai
Golestaneh, S. Alireza
Chandra, Shubhra
Zhou, Min
contents We present ALF (Advertiser Large Foundation model), a multi-modal transformer architecture for understanding advertiser behavior and intent across text, image, video, and structured data modalities. Through contrastive learning and multi-task optimization, ALF creates unified advertiser representations that capture both content and behavioral patterns. Our model achieves state-of-the-art performance on critical tasks including fraud detection, policy violation identification, and advertiser similarity matching. In production deployment, ALF demonstrates significant real-world impact by delivering simultaneous gains in both precision and recall, for instance boosting recall by over 40 percentage points on one critical policy and increasing precision to 99.8% on another. The architecture's effectiveness stems from its novel combination of multi-modal transformations, inter-sample attention mechanism, spectrally normalized projections, and calibrated probabilistic outputs.
format Preprint
id arxiv_https___arxiv_org_abs_2504_18785
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ALF: Advertiser Large Foundation Model for Multi-Modal Advertiser Understanding
Rajagopalan, Santosh
Vronsky, Jonathan
Yan, Songbai
Golestaneh, S. Alireza
Chandra, Shubhra
Zhou, Min
Machine Learning
We present ALF (Advertiser Large Foundation model), a multi-modal transformer architecture for understanding advertiser behavior and intent across text, image, video, and structured data modalities. Through contrastive learning and multi-task optimization, ALF creates unified advertiser representations that capture both content and behavioral patterns. Our model achieves state-of-the-art performance on critical tasks including fraud detection, policy violation identification, and advertiser similarity matching. In production deployment, ALF demonstrates significant real-world impact by delivering simultaneous gains in both precision and recall, for instance boosting recall by over 40 percentage points on one critical policy and increasing precision to 99.8% on another. The architecture's effectiveness stems from its novel combination of multi-modal transformations, inter-sample attention mechanism, spectrally normalized projections, and calibrated probabilistic outputs.
title ALF: Advertiser Large Foundation Model for Multi-Modal Advertiser Understanding
topic Machine Learning
url https://arxiv.org/abs/2504.18785