AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Parameter and Computation Efficient Transfer Learning for Vision-Language Pre-trained Models

Qiong Wu, Wei Yu, Yiyi Zhou, Shubin Huang, Xiaoshuai Sun, Rongrong Ji

NeurIPS 2023 · Architectures and efficiency › Pruning and efficient tuning · approach: Efficient ViT adaptation · 4 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
skip 8 fusion layers instead of 0 with DAS-Fusion (ablation)DAS-Fusion · VQA test-dev (METER)accuracy: 75.28 -> 73.70Mixed
skip 8 fusion layers instead of 0 with DAS-Fusion (ablation)DAS-Fusion · VQA additional FLOPs (METER)additional FLOPs: 1.68G -> -24.00GImproved
skip 8 layers instead of 0 with DAS-Global (ablation)DAS-Global · NLVR2 test-P (METER)accuracy: 81.28 -> 79.61Hurt
skip 2 layers instead of 1 with DAS on ViLT (ablation)DAS · VQA test-dev (ViLT)accuracy: 69.28 -> 67.64Hurt
skip 1 layer with DAS on ViLT vs no skipping (ablation)DAS · VQA test-dev (ViLT)accuracy: 70.13 -> 69.28Hurt
skip fusion layers only (DAS-Fusion) instead of whole model (DAS-Global), 4 skipped (ablation)DAS · VQA additional FLOPs (METER)additional FLOPs: -4.51G -> -11.16GImproved
RL-based DAS search instead of random skipping (ablation)DAS · VQA2.0 (METER)accuracy: +-1.33 deviation for random skipping of 8 layersImproved
DAS4-Fusion instead of full tuning for inference speedDAS4-Fusion · VQA (METER)test speed (samples/s): 4.16 -> 4.96Improved
DAS4-Global instead of full tuning for inference speedDAS4-Global · VQA (METER)test speed (samples/s): 4.16 -> 4.57Improved
Scaled PA instead of full tuning for inference speedScaled PA · VQA (METER)test speed (samples/s): 4.16 -> 3.95Hurt
DAS4-Fusion instead of Scaled PADAS4-Fusion · VQA test-dev (METER)accuracy: 75.11 -> 74.80No clear effect
LoRA instead of AdapterLoRA · VQA test-dev (METER)accuracy: 74.70 -> 74.00Hurt

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