I3: Intent-Introspective Retrieval Conditioned on Instructions
Kaihang Pan, Juncheng Li, Wenjie Wang, Hao Fei, Hongye Song, Wei Ji,, Jun Lin, Xiaozhong Liu, Tat-Seng Chua, Siliang Tang

TL;DR
I3 is a unified retrieval system that uses instructions and an introspector to perform intent-aware retrieval across various tasks without task-specific training, achieving state-of-the-art zero-shot results.
Contribution
The paper introduces I3, a novel intent-introspective retrieval framework that leverages instructions and a pluggable introspector for zero-shot, intent-aware retrieval across diverse tasks.
Findings
Outperforms task-specific retrievers on BEIR benchmark
Achieves state-of-the-art zero-shot retrieval performance
Effectively incorporates intent understanding without task-specific training
Abstract
Recent studies indicate that dense retrieval models struggle to perform well on a wide variety of retrieval tasks that lack dedicated training data, as different retrieval tasks often entail distinct search intents. To address this challenge, in this work we leverage instructions to flexibly describe retrieval intents and introduce I3, a unified retrieval system that performs Intent-Introspective retrieval across various tasks, conditioned on Instructions without any task-specific training. I3 innovatively incorporates a pluggable introspector in a parameter-isolated manner to comprehend specific retrieval intents by jointly reasoning over the input query and instruction, and seamlessly integrates the introspected intent into the original retrieval model for intent-aware retrieval. Furthermore, we propose progressively-pruned intent learning. It utilizes extensive LLM-generated data to…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsMultimodal Machine Learning Applications · Domain Adaptation and Few-Shot Learning · Advanced Image and Video Retrieval Techniques
MethodsPruning
