AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

FlowCIR: Semantic Transport via Flow Matching for Zero-Shot Composed Image Retrieval

Zhenqi He, Ziqi Jiang, Yuanpei Liu, Yanghao Wang, Teng Wang, Long Chen

ECCV 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Attribute-based zero-shot learning · 2 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 conditional flow matching on top of text-only query (ablation)FlowCIR (ViT-L/14) · CIRCO testmAP@5: 3.3 -> 13.5Improved
add conditional flow matching on top of text-only query (ablation)FlowCIR (ViT-L/14) · CIRR testR@5: 44.0 -> 53.8Improved
add Multi-Negative Steering on top of CFM (ablation)FlowCIR (ViT-L/14) · CIRCO testmAP@5: 13.5 -> 14.9Improved
add Multi-Negative Steering on top of CFM (ablation)FlowCIR (ViT-L/14) · CIRR testR@5: 53.8 -> 56.1Improved
add Multi-Negative Steering to text-only baseline (no CFM) (ablation)Text-only query · CIRR and CIRCO testR@5 / mAP@5: CIRR R@5 44.0 -> 45.7; CIRCO mAP@5 3.3 -> 2.5Mixed
replace flow matching with direct regression of target residual (ablation)FlowCIR · CIRCO testmAP@5: 14.9 -> 12.3Hurt
replace flow matching with direct regression of target residual (ablation)FlowCIR · CIRR testR@1: 26.2 -> 23.8Hurt
transport toward target image embedding instead of text embedding (ablation)FlowCIR · CIRCO testmAP@5: 14.9 -> 12.7Hurt
transport toward target image embedding instead of text embedding (ablation)FlowCIR · CIRR testR@1: 26.2 -> 23.8Hurt
raise retrieval loss weight lambda from 0.1 to 0.5 (ablation)FlowCIR · CIRCO testmAP@5: 10.6 -> 14.9Improved
raise retrieval loss weight lambda from 0.5 to 0.9 (ablation)FlowCIR · CIRR/CIRCO testR@1 / mAP@5: CIRR R@1 26.2 -> 25.8; CIRCO mAP@5 14.9 -> 14.8No clear effect
use hard-negative count K=B/8 instead of B/4 or B/16 (ablation)FlowCIR · CIRCO testmAP@5: 13.7 (B/4), 14.2 (B/16) -> 14.9 (B/8)Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 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