AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

INTENT: Invariance and Discrimination-aware Noise Mitigation for Robust Composed Image Retrieval

Zhiwei Chen, Yupeng Hu, Zhiheng Fu, Zixu Li, Jiale Huang, Qinlei Huang, Yinwei Wei

AAAI 2026 · Vision-language and multimodal › Cross-modal learning · approach: Probabilistic embeddings for cross-modal retrieval · 11 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add Visual Invariant Composition (FFT counterfactual consistency) (ablation)INTENT · FashionIQ-Avg R@10, 20% noiseR@10: 53.05 -> 54.22Improved
use FFT counterfactual images instead of grayscale images (ablation)INTENT · FashionIQ-Avg R@10, 20% noiseR@10: 54.01 -> 54.22Improved
add positive weight reward to decision boundary (ablation)INTENT · FashionIQ-Avg R@10, 20% noiseR@10: 53.37 -> 54.22Improved
add negative weight reward to decision boundary (ablation)INTENT · FashionIQ-Avg R@10, 20% noiseR@10: 53.89 -> 54.22Improved
add both rewards (scalable decision boundary) (ablation)INTENT · CIRR-Avg R@K, 20% noiseR@K: 77.93 -> 80.17Improved
add robust contrastive loss (ablation)INTENT · CIRR-Avg R@K, 20% noiseR@K: 78.33 -> 80.17Improved
mask positives in the robust contrastive loss (ablation)INTENT · CIRR-Avg R@K, 20% noiseR@K: 74.26 -> 80.17Improved
add soft discriminative loss (ablation)INTENT · CIRR-Avg R@K, 20% noiseR@K: 77.67 -> 80.17Improved
use CKA instead of MSE for causal consistency loss (ablation)INTENT · FashionIQ-Avg R@10, 20% noiseR@10: 53.88 -> 54.22Improved
INTENT vs TME under high noiseINTENT · FashionIQ val, 80% noiseAVG: 57.63 -> 59.07Improved
INTENT vs TME under no noiseINTENT · CIRR test, 0% noiseAvg(R@5, Rsub@1): 82.01 -> 81.70No clear effect
INTENT vs TME at 20% noiseINTENT · CIRR test, 20% noiseAvg(R@5, Rsub@1): 79.74 -> 79.66No clear effect

Later papers that compared with this method

1 later accepted paper reports INTENT as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy