AI Paper Atlas

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

Auto-Prompting: Layer-Specific Prompt Fusion Discovery via Differentiable Search

Xi Xiao, Xingjian Li, Yunbei Zhang, Cheng Han, Tianming Liu, Tianyang Wang, Runmin Jiang, Jihun Hamm, Xiao Wang, Min Xu

ECCV 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Prompt learning for CLIP

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
replace fixed cross-attention fusion with searched layer-wise hybrid fusion (ablation)VPT with frozen ViT-B/16 · VTAB-1k Naturalaccuracy: 79.91 -> 82.88Improved
replace fixed cross-attention with searched fusion (ablation)VPT with frozen ViT-B/16 · VTAB-1k Structuredaccuracy: 58.67 -> 62.55Improved
replace fixed concat with searched fusion (ablation)VPT with frozen ViT-B/16 · VTAB-1k Naturalaccuracy: 78.42 -> 82.88Improved
replace fixed add with searched fusion (ablation)VPT with frozen ViT-B/16 · VTAB-1k Naturalaccuracy: 77.63 -> 82.88Improved
replace fixed affine with searched fusion (ablation)VPT with frozen ViT-B/16 · VTAB-1k Structuredaccuracy: 57.40 -> 62.55Improved
use searched fusion instead of VPT-Deep concatenationViT-B/16 · VTAB-1k meanaccuracy: 69.43 -> 77.01Improved
searched fusion under MAE pretraining vs VFPTViT-B/16 MAE · VTAB-1k Naturalaccuracy: 53.59 -> 55.12Improved
searched fusion under MAE pretraining vs Partial-1 / BiasViT-B/16 MAE · VTAB-1kaccuracy: 39.01 vs 47.64 Structured (Partial-1)Mixed
searched fusion under MoCo v3ViT-B/16 MoCo v3 · VTAB-1k Naturalaccuracy: 77.47 -> 79.60Improved
searched fusion on Swin-Base vs VPT-DeepSwin-Base · VTAB-1k Structuredaccuracy: +10.51Improved
searched fusion adds inference costViT-B/16 · VTAB-1klatency per image: 14.8 -> 15.9 msHurt
searched fusion adds search costViT-B/16 · VTAB-1ktraining GPU hours: 7.8 -> 10.8 GPU hoursHurt

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