AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

A-La-Carte Prompt Tuning (APT): Combining Distinct Data via Composable Prompting

Benjamin Bowman, Alessandro Achille, Luca Zancato, Matthew Trager, Pramuditha Perera, Giovanni Paolini, Stefano Soatto

CVPR 2023 · Architectures and efficiency › Pruning and efficient tuning · approach: Low-rank compression and merging · 10 citations (OpenAlex)

Read the paper

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
Compose prompts trained independently on data shards using structured attentionAPT · Image classification datasets, including 20-shard settingsaccuracy: APT accuracy at 20 shards: 86.3%; 83.9%; 88.7%; 93.3%; 45.4%; 97.6%; 45.8%; full-data accuracy: 86.2%; 86.6%; 91.7%; 93.Mixed
Add prototype-distance weighting to prompt logits (ablation)APT · Split CIFAR-100average accuracy: 83.63 -> 85.21Improved
Add prototype-distance weighting to prompt logits (ablation)APT · CORe50average accuracy: 90.89 -> 91.14Improved
Use structured attention rather than naïvely concatenating prompts (ablation)APT · Prompt composition experimentsaccuracyImproved
Use prompt tuning rather than training only classifier heads (ablation)APT shard composition · In-domain and out-of-domain datasetsaccuracy: in-domain, 20 shards: 90.0% vs. 88.5%; out-of-domain, 20 shards: 45.6% vs. 40.5%Improved
Use average ensembling rather than majority vote (ablation)APT prompt predictions · Shard ensemble datasetsaccuracy: Average ensembling minus majority vote: 3.81% (2 shards); average is reported in the range 1.5-3.8%.Improved
Add per-layer memory tokens instead of using a single shallow prompt without memory (ablation)Prompt tuning · MIT-67accuracy: 86.0% -> 86.2%Improved
Use per-layer memory tokens rather than sharing memory tokens across layers (ablation)Prompt tuning · MIT-67accuracy: 86.5% -> 86.2%Hurt
Replace ImageNet-21k pretraining with the ALBEF visual encoder (ablation)APT · Image classification datasets, 20 shardsaverage accuracy: 77.3% -> 67.34%Hurt
Use a fixed pretrained backbone and prompts rather than fine-tuning separate models per shardShard ensembles · All seven image classification datasets, 20 shardsaverage accuracy: 41.46% -> 77.3%Improved
Increase the number of data shards from 2 to 20APT · Average across seven image classification datasetsaccuracy: 84.2% -> 77.3%Hurt

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