AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Heterogeneous Uncertainty-Guided Composed Image Retrieval with Fine-Grained Probabilistic Learning

Haomiao Tang, Jinpeng Wang, Minyi Zhao, GuangHao Meng, Ruisheng Luo, Long Chen, Shu-Tao Xia

AAAI 2026 · Vision-language and multimodal › Human and embodied modeling · approach: Composed and text-based image retrieval · 3 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 probabilistic embeddings to point matching (ablation)HUG · Fashion-IQ valaverage recall: 52.26 -> 55.44Improved
add component-wise fine-grained contrast (ablation)HUG · Fashion-IQ valaverage recall: 55.44 -> 57.97Improved
add instance-wise fine-grained contrast (ablation)HUG · Fashion-IQ valaverage recall: 57.97 -> 58.48Improved
add modality-wise fine-grained contrast (ablation)HUG · Fashion-IQ valaverage recall: 58.48 -> 59.33Improved
add naive cross-modal uncertainty (ablation)HUG · Fashion-IQ valaverage recall: 59.33 -> 58.57Hurt
add multi-modal coordination loss (ablation)HUG · Fashion-IQ valaverage recall: 58.57 -> 63.11Improved
add dynamic weighting instead of static averaging (ablation)HUG · Fashion-IQ valaverage recall: 63.11 -> 63.74Improved
add probabilistic uncertainty modelling cost (ablation)HUG · Fashion-IQ valavg time per query (ms): 7.51 -> 21.35Hurt
set lambda_Cord to 0.1 rather than larger (ablation)HUG · Fashion-IQ valaverage recallImproved
reduce lambda_FC (ablation)HUG · Fashion-IQ valaverage recallHurt
use HUG instead of QuReHUG · CIRR testRecall@1: 52.22 vs 51.09Mixed
use HUG instead of QuReHUG · Fashion-IQ valaverage recall: 63.04 -> 63.74Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

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