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

arXiv: https://arxiv.org/abs/2604.19386 · HTML: https://mlsanity.com/paper/2604.19386/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- Builds on 85 of its 133 references that are in the map.
- First posted on arXiv 21 Apr 2026.

## Builds on (most cited first, 30 of 85)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?](https://mlsanity.com/paper/1703.04977/) (2017)
- [Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labels](https://mlsanity.com/paper/1804.06872/) (2018)
- [A Closer Look at Memorization in Deep Networks](https://mlsanity.com/paper/1706.05394/) (2017)
- [CogVLM: Visual Expert for Pretrained Language Models](https://mlsanity.com/paper/2311.03079/) (2023)
- [A Simple Baseline for Bayesian Uncertainty in Deep Learning](https://mlsanity.com/paper/1902.02476/) (2019)
- [Laplace Redux -- Effortless Bayesian Deep Learning](https://mlsanity.com/paper/2106.14806/) (2021)
- [Unsupervised Label Noise Modeling and Loss Correction](https://mlsanity.com/paper/1904.11238/) (2019)
- [Image Retrieval on Real-life Images with Pre-trained Vision-and-Language Models](https://mlsanity.com/paper/2108.04024/) (2021)
- [FutureSightDrive: Thinking Visually with Spatio-Temporal CoT for Autonomous Driving](https://mlsanity.com/paper/2505.17685/) (2025)
- [JanusVLN: Decoupling Semantics and Spatiality with Dual Implicit Memory for Vision-Language Navigation](https://mlsanity.com/paper/2509.22548/) (2025)
- [Transformer Tracking with Cyclic Shifting Window Attention](https://mlsanity.com/paper/2205.03806/) (2022)
- [SE-Agent: Self-Evolution Trajectory Optimization in Multi-Step Reasoning with LLM-Based Agents](https://mlsanity.com/paper/2508.02085/) (2025)
- [PRISM: Self-Pruning Intrinsic Selection Method for Training-Free Multimodal Data Selection](https://mlsanity.com/paper/2502.12119/) (2025)
- [Sentence-level Prompts Benefit Composed Image Retrieval](https://mlsanity.com/paper/2310.05473/) (2023)
- [LLaVA Steering: Visual Instruction Tuning with 500x Fewer Parameters through Modality Linear Representation-Steering](https://mlsanity.com/paper/2412.12359/) (2024)
- [CoT-Kinetics: A Theoretical Modeling Assessing LRM Reasoning Process](https://mlsanity.com/paper/2505.13408/) (2025)
- [Compact Transformer Tracker with Correlative Masked Modeling](https://mlsanity.com/paper/2301.10938/) (2023)
- [FineCIR: Explicit Parsing of Fine-Grained Modification Semantics for Composed Image Retrieval](https://mlsanity.com/paper/2503.21309/) (2025)
- [ReTrack: Evidence-Driven Dual-Stream Directional Anchor Calibration Network for Composed Video Retrieval](https://mlsanity.com/paper/2604.17898/) (2026)
- [OFFSET: Segmentation-based Focus Shift Revision for Composed Image Retrieval](https://mlsanity.com/paper/2507.05631/) (2025)
- [ASCD: Attention-Steerable Contrastive Decoding for Reducing Hallucination in MLLM](https://mlsanity.com/paper/2506.14766/) (2025)
- [Chat-Driven Text Generation and Interaction for Person Retrieval](https://mlsanity.com/paper/2509.12662/) (2025)
- [Curriculum-RLAIF: Curriculum Alignment with Reinforcement Learning from AI Feedback](https://mlsanity.com/paper/2505.20075/) (2025)
- [HUD: Hierarchical Uncertainty-Aware Disambiguation Network for Composed Video Retrieval](https://mlsanity.com/paper/2512.02792/) (2025)
- [Coupled Mamba: Enhanced Multi-modal Fusion with Coupled State Space Model](https://mlsanity.com/paper/2405.18014/) (2024)
- [Prior Knowledge Integration via LLM Encoding and Pseudo Event Regulation for Video Moment Retrieval](https://mlsanity.com/paper/2407.15051/) (2024)

## Built on by (most cited first, 18 of 18)

- [ConeSep: Cone-based Robust Noise-Unlearning Compositional Network for Composed Image Retrieval](https://mlsanity.com/paper/2604.20358/) (2026)
- [Bridging Reasoning Trajectories in On-Policy Distillation via Near-Future Guidance](https://mlsanity.com/paper/2606.00305/) (2026)
- [Cornerstones or Stumbling Blocks? Deciphering the Rock Tokens in On-Policy Distillation](https://mlsanity.com/paper/2605.09253/) (2026)
- [COMBINER: Composed Image Retrieval Guided by Attribute-based Neighbor Relations](https://mlsanity.com/paper/2606.04604/) (2026)
- [EgoAction: Egocentric Action Composition with Reliability-Aware Temporal Fusion for the EPIC-KITCHENS Action Detection Challenge at CVPR 2026](https://mlsanity.com/paper/2605.24496/) (2026)
- [EgoAdapt: A Multi-Scene Egocentric Adaptation Method for CVPR 2026 HD-EPIC VQA Challenge](https://mlsanity.com/paper/2605.24500/) (2026)
- [TempRet: Temporal Enhancement and Two-Stage Reranking for CVPR 2026 EPIC-KITCHENS-100 Multi-Instance Retrieval Challenge](https://mlsanity.com/paper/2605.24470/) (2026)
- [OmniEgo-R2: A Routed Reasoning Framework for the 1st Cross-Domain EgoCross Challenge at CVPR 2026](https://mlsanity.com/paper/2605.24481/) (2026)
- [R^3: Composed Video Retrieval via Reasoning-Guided Recalling and Re-ranking](https://mlsanity.com/paper/2606.01113/) (2026)
- [OmniTrend: Content-Context Modeling for Scalable Social Popularity Prediction](https://mlsanity.com/paper/2604.26252/) (2026)
- [L-Drive: Beyond a Single Mapping-Latent Context Drives Time Series Forecasting](https://mlsanity.com/paper/2605.17730/) (2026)
- [Conformal Risk Prediction for Non-Alcoholic Fatty Liver Disease Using Gradient Boosting with Distribution-Free Coverages](https://mlsanity.com/paper/2606.09860/) (2026)
- [On the Controllability-Fidelity Frontier in Diffusion Editing](https://mlsanity.com/paper/2606.09901/) (2026)
- [VigilFormer: Deformable Attention for Video Anomaly Detection with Causal Risk Inference](https://mlsanity.com/paper/2606.14724/) (2026)
- [URDF Synthesis from RGB-D Sequences via Differentiable Joint Inference and Energy-Consistent Verification](https://mlsanity.com/paper/2606.18861/) (2026)
- [MOSAIC: Orchestrating Collaborative Knowledge Tracing with Hierarchical Semantic Alignment](https://mlsanity.com/paper/2606.29049/) (2026)
- [Cross-Modal Hierarchical Fusion for from Multi-Sensor Ground Observation](https://mlsanity.com/paper/2606.30647/) (2026)
- [From Transformation to Target State: Rethinking Query Representation for Zero-Shot Composed Image Retrieval](https://mlsanity.com/paper/2610.05993/) (2026)

## Abstract (arXiv)

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.