Paper Image

Multimodal image search query suggestions

Published on:

7 February 2024

Primary Category:

Information Retrieval

Paper Authors:

Zheng Wang,

Bingzheng Gan,

Wei Shi

Bullets

Key Details

Presents RL4Sugg system for multimodal query suggestions from images

Uses large language models and multi-agent reinforcement learning

Training incorporates human feedback signals

Achieved 18% better performance than previous approaches

Implemented in search engines to improve user experience

AI generated summary

Multimodal image search query suggestions

This paper introduces a system called RL4Sugg that generates query suggestions based on user images to improve search intent and result diversity. It uses large language models and multi-agent reinforcement learning optimized by human feedback. Experiments showed an 18% improvement over existing methods. The system was added to search engines and increased user engagement.

Answers from this paper

Comments

No comments yet, be the first to start the conversation...

Sign up to comment on this paper

Sign Up