Air-Know: Arbiter-Calibrated Knowledge-Internalizing Robust Network for Composed Image Retrieval

2026-04-21

AI Summary

This work addresses the semantic ambiguity and representation corruption caused by noisy triplet correspondences (NTC) in compositional image retrieval by proposing a novel “expert-agent-split” decoupling paradigm. The method introduces a high-precision offline anchor set constructed via multimodal large language models as an external expert, whose discriminative logic is internalized through a lightweight proxy arbitrator. Coupled with a dual-stream architecture and a matching-confidence-driven data splitting strategy, this approach effectively breaks the self-referential loop between the learner and the arbitrator. Extensive experiments demonstrate that the proposed method significantly outperforms existing state-of-the-art approaches across multiple CIR benchmarks, exhibiting particularly pronounced advantages under NTC settings while maintaining strong competitiveness in conventional retrieval tasks.

Technology Category

• Computer Vision: Image and Video Retrieval
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Composed Image Retrieval (CIR) has attracted significant attention due to its flexible multimodal query method, yet its development is severely constrained by the Noisy Triplet Correspondence (NTC) problem. Most existing robust learning methods rely on the "small loss hypothesis", but the unique semantic ambiguity in NTC, such as "partial matching", invalidates this assumption, leading to unreliable noise identification. This entraps the model in a self dependent vicious cycle where the learner is intertwined with the arbiter, ultimately causing catastrophic "representation pollution". To address this critical challenge, we propose a novel "Expert-Proxy-Diversion" decoupling paradigm, named Air-Know (ArbIteR calibrated Knowledge iNternalizing rObust netWork). Air-Know incorporates three core modules: (1) External Prior Arbitration (EPA), which utilizes Multimodal Large Language Models (MLLMs) as an offline expert to construct a high precision anchor dataset; (2) Expert Knowledge Internalization (EKI), which efficiently guides a lightweight proxy "arbiter" to internalize the expert's discriminative logic; (3) Dual Stream Reconciliation (DSR), which leverages the EKI's matching confidence to divert the training data, achieving a clean alignment stream and a representation feedback reconciliation stream. Extensive experiments on multiple CIR benchmark datasets demonstrate that Air-Know significantly outperforms existing SOTA methods under the NTC setting, while also showing strong competitiveness in traditional CIR.

Problem

• Composed Image Retrieval
• Noisy Triplet Correspondence
• Semantic Ambiguity
• Representation Pollution
• Robust Learning

Innovation

• Composed Image Retrieval
• Noisy Triplet Correspondence
• Multimodal Large Language Models
• Knowledge Internalization
• Decoupling Paradigm