AI Paper Atlas

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

Bridging Vision and Language Encoders: Parameter-Efficient Tuning for Referring Image Segmentation

Zunnan Xu, Zhihong Chen, Yong Zhang, Yibing Song, Xiang Wan, Guanbin Li

ICCV 2023 · Vision-language and multimodal › CLIP and zero-shot · approach: Parameter-efficient visual tuning · 99 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 Bridger to frozen CLIP backbone (ablation)ETRIS (ResNet-101) · RefCOCO valoIoU: 67.73 -> 71.06Improved
add Hierarchical Alignment Module (ablation)ETRIS (ResNet-101) · RefCOCO valoIoU: 58.70 -> 71.06Improved
add Global Alignment Module (ablation)ETRIS (ResNet-101) · RefCOCO valoIoU: 62.21 -> 71.06Improved
expand Bridger fusion scope from n->n to 2->n (ablation)ETRIS (ResNet-101) · RefCOCO valoIoU: 69.65 -> 70.96Improved
increase number of Bridgers from 1 to 3 (ablation)ETRIS scope 2->n · RefCOCO valoIoU: 70.96 -> 71.06No clear effect
use Conv&Deconv zoom layer instead of Linear, MLP or Conv&Interpolate (ablation)ETRIS (ResNet-101) · RefCOCO valoIoU: 70.08 -> 71.06Improved
increase Bridger hidden dim from 64 to 128 (ablation)ETRIS (ResNet-101) · RefCOCO valoIoU: 71.06 -> 70.39Hurt
use Bridger instead of Adapter/LoRA/Compacter/CoOp as PET methodfrozen CLIP ResNet-101 · RefCOCO valoIoU: 69.46 -> 71.06Improved
combine Conv Adapter with BridgerETRIS (ResNet-101) · RefCOCO valoIoU: 71.06 -> 72.11Improved
tune only ~1.9M backbone params with Bridger instead of full fine-tuningCLIP ResNet-101 · RefCOCO valoIoU: 70.47 -> 71.06Improved

Earlier methods this paper compares with

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