AI Paper Atlas

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

Exploring Cross-Modal Flows for Few-Shot Learning

Ziqi Jiang, Yanghao Wang, Long Chen

ICLR 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Parameter-efficient visual tuning

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 FMA multi-step flow-matching alignment on top of CLIP-LoRA (16-shot)CLIP-LoRA · difficult datasets avg, 16-shotaccuracy: 76.1 -> 77.8Improved
add FMA on zero-shot CLIP features (16-shot)CLIP · difficult datasets avgaccuracy: 48.9 -> 68.9 (+20.0)Improved
add FMA on CoOpCoOp · difficult datasets avg, 16-shotaccuracy: 71.4 -> 74.0Improved
add FMA on CoCoOpCoCoOp · difficult datasets avg, 16-shotaccuracy: 64.1 -> 68.5Improved
add FMA on CLIP-AdapterCLIP-Adapter · difficult datasets avg, 16-shotaccuracy: 62.4 -> 67.9Improved
apply FMA to easy datasets rather than difficult onesCLIP-LoRA · 16-shot easy datasets avgaccuracy: 88.2 -> 88.8Improved
add FMA when fewer shots (1-shot) on CLIP-LoRACLIP-LoRA · 1-shot difficult avgaccuracy: 58.7 -> 59.4Improved
add FMA trained on ImageNet, evaluated cross-datasetCoOp · generalization on easy datasetsaccuracy: 75.8 -> 75.7No clear effect
add FMA, harmonic mean of adaptation and generalizationCLIP · difficult datasetsharmonic mean accuracy: 48.9 -> 57.5Improved
use early stopping solver instead of vanilla flow-matching solver (ablation)FMA · DTD 16-shotaccuracyImproved
increase inference steps beyond about 8 with early stopping solver (ablation)FMA · DTD 16-shotaccuracyHurt
add noise augmentation during velocity training (ablation)FMA on CLIP-LoRA · difficult datasetsaccuracyImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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