ConeSep: Cone-based Robust Noise-Unlearning Compositional Network for Composed Image Retrieval

2026-04-22

AI Summary

This work addresses the “hard noise” problem in compositional image retrieval caused by erroneous triplet annotations, which violates the conventional small-loss assumption and leads to overlooked challenges such as modality suppression, missing negative anchors, and forgetting backlash. To tackle these issues, the authors propose a robust denoising learning framework grounded in conic geometry. The method precisely identifies noisy samples through geometric fidelity quantification, constructs semantically adversarial anchor points via negative boundary learning, and incorporates an optimal transport–based directional forgetting mechanism to rectify noise. Extensive experiments on the FashionIQ and CIRR benchmarks demonstrate that the proposed approach significantly outperforms current state-of-the-art methods, confirming its effectiveness and robustness.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Adversarial Learning & Robustness
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

The Composed Image Retrieval (CIR) task provides a flexible retrieval paradigm via a reference image and modification text, but it heavily relies on expensive and error-prone triplet annotations. This paper systematically investigates the Noisy Triplet Correspondence (NTC) problem introduced by annotations. We find that NTC noise, particularly ``hard noise'' (i.e., the reference and target images are highly similar but the modification text is incorrect), poses a unique challenge to existing Noise Correspondence Learning (NCL) methods because it breaks the traditional ``small loss hypothesis''. We identify and elucidate three key, yet overlooked, challenges in the NTC task, namely (C1) Modality Suppression, (C2) Negative Anchor Deficiency, and (C3) Unlearning Backlash. To address these challenges, we propose a Cone-based robuSt noisE-unlearning comPositional network (ConeSep). Specifically, we first propose Geometric Fidelity Quantization, theoretically establishing and practically estimating a noise boundary to precisely locate noisy correspondence. Next, we introduce Negative Boundary Learning, which learns a ``diagonal negative combination'' for each query as its explicit semantic opposite-anchor in the embedding space. Finally, we design Boundary-based Targeted Unlearning, which models the noisy correction process as an optimal transport problem, elegantly avoiding Unlearning Backlash. Extensive experiments on benchmark datasets (FashionIQ and CIRR) demonstrate that ConeSep significantly outperforms current state-of-the-art methods, which fully demonstrates the effectiveness and robustness of our method.

Problem

• Composed Image Retrieval
• Noisy Triplet Correspondence
• Hard Noise
• Noise Correspondence Learning
• Annotation Noise

Innovation

• Composed Image Retrieval
• Noisy Triplet Correspondence
• Noise-Unlearning
• Negative Anchor
• Optimal Transport