LLMDistill4Ads: Using Cross-Encoders to Distill from LLM Signals for Advertiser Keyphrase Recommendations at eBay

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Dey, Soumik, Braun, Benjamin, Ravipati, Naveen, Wu, Hansi, Li, Binbin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917280309313536
author Dey, Soumik
Braun, Benjamin
Ravipati, Naveen
Wu, Hansi
Li, Binbin
author_facet Dey, Soumik
Braun, Benjamin
Ravipati, Naveen
Wu, Hansi
Li, Binbin
contents E-commerce sellers are advised to bid on keyphrases to boost their advertising campaigns. These keyphrases must be relevant to prevent irrelevant items from cluttering Search systems and to maintain positive seller perception. It is vital that keyphrase suggestions align with seller, Search, and buyer judgments. Given the challenges in collecting negative feedback in these systems, LLMs have been used as a scalable proxy for human judgments. We present an empirical study on a major e-commerce platform of a distillation framework involving an LLM teacher, a cross-encoder assistant and a bi-encoder Embedding Based Retrieval (EBR) student model, aimed at mitigating click-induced biases and provide more diverse keyphrase recommendations while aligning advertising, search and buyer preferences.
format Preprint
id arxiv_https___arxiv_org_abs_2508_03628
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LLMDistill4Ads: Using Cross-Encoders to Distill from LLM Signals for Advertiser Keyphrase Recommendations at eBay
Dey, Soumik
Braun, Benjamin
Ravipati, Naveen
Wu, Hansi
Li, Binbin
Information Retrieval
Artificial Intelligence
Machine Learning
E-commerce sellers are advised to bid on keyphrases to boost their advertising campaigns. These keyphrases must be relevant to prevent irrelevant items from cluttering Search systems and to maintain positive seller perception. It is vital that keyphrase suggestions align with seller, Search, and buyer judgments. Given the challenges in collecting negative feedback in these systems, LLMs have been used as a scalable proxy for human judgments. We present an empirical study on a major e-commerce platform of a distillation framework involving an LLM teacher, a cross-encoder assistant and a bi-encoder Embedding Based Retrieval (EBR) student model, aimed at mitigating click-induced biases and provide more diverse keyphrase recommendations while aligning advertising, search and buyer preferences.
title LLMDistill4Ads: Using Cross-Encoders to Distill from LLM Signals for Advertiser Keyphrase Recommendations at eBay
topic Information Retrieval
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2508.03628