AI Paper Atlas

Research map / Continual learning / Online and replay-based

AttriCLIP: A Non-Incremental Learner for Incremental Knowledge Learning

Runqi Wang, Xiaoyue Duan, Guoliang Kang, Jianzhuang Liu, Shaohui Lin, Songcen Xu, Jinhu Lü, Baochang Zhang

CVPR 2023 · Continual learning › Online and replay-based · approach: Dynamic expansion and isolation · 48 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
use attribute-keyed prompts rather than category-specific prompt learningAttriCLIP · CIFAR100, Task10average accuracy: 79.7 vs 66.7Improved
use attribute-keyed prompts rather than category-specific prompt learningAttriCLIP · ImageNet100, Task10average accuracy: 81.4 vs 75.4Improved
transfer learned attribute prompts from ImageNet100 before training on CIFAR100AttriCLIP · CIFAR100-I2Cforward transfer: +0.9Improved
transfer learned attribute prompts from CIFAR100 to ImageNet100AttriCLIP · ImageNet100-I2Cbackward transfer: +7.0Improved
train with attribute-keyed prompts for the sequential ImageNet100 and CIFAR100 settingAttriCLIP · CIFAR100 + ImageNet100 CDCLaccuracy: 78.3 vs 67.1Improved
use cosine distance rather than triplet loss for key matching (ablation)AttriCLIP · CIFAR100average accuracy: 80.22 -> 81.38Improved
use cosine distance rather than MSE for key matching (ablation)AttriCLIP · CIFAR100average accuracy: 80.81 -> 81.38Improved
add prompt diversity loss with λp=0.3 instead of omitting it (ablation)AttriCLIP · CIFAR100average accuracy: 78.10 -> 81.38Improved
set key matching loss weight λk to 0.7 (ablation)AttriCLIP · CIFAR100average accuracy: 80.28 -> 81.38Improved
vary prompt length, attribute-bank size, and selected-key count (ablation)AttriCLIP · CIFAR100average accuracyMixed

Later papers that compared with this method

2 later accepted papers report AttriCLIP as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

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