AI Paper Atlas

Research map / Continual learning / Class-incremental

External Knowledge Injection for CLIP-Based Class-Incremental Learning

Da-Wei Zhou, Kai-Wen Li, Jingyi Ning, Han-Jia Ye, Lijun Zhang, De-Chuan Zhan

ICCV 2025 · Continual learning › Class-incremental · approach: Exemplar-efficient class-incremental learning · 22 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 visual knowledge injection (augmentation-based unit) to zero-shot CLIP (ablation)ENGINE · CIFAR100 B0 Inc10accuracyImproved
add textual (GPT-4 description) injection on top of visual injection (ablation)ENGINE · CIFAR100 B0 Inc10accuracyImproved
add post-tuning knowledge injection (top-k pairwise re-ranking) (ablation)ENGINE · CIFAR100 B0 Inc10accuracyImproved
vary prototype noise alpha in {0.15..0.35} and top-k in {3..7} (ablation)ENGINE · CUB B0 Inc20last accuracyNo clear effect
replace GPT-4o mini descriptions with Qwen2.5-72B (ablation)ENGINE · Food B0 Inc10accuracyNo clear effect
give baselines the same GPT-generated text and augmentation as ENGINE (ablation)RAPF vs ENGINE · ImageNet-R B0 Inc20average accuracy: 81.89 -> 86.22Improved
give ZS-CLIP the same GPT-generated text (ablation)ZS-CLIP vs ENGINE · CIFAR B0 Inc10average accuracy: 82.26 -> 86.92Improved
ENGINE without exemplars versus exemplar-based PROOF (20/class)ENGINE · Aircraft B0 Inc10average accuracy: 61.00 -> 69.69Improved
ENGINE without exemplars versus exemplar-based iCaRL (20/class)ENGINE · SUN B0 Inc30average accuracy: 78.56 -> 85.04Improved
ENGINE versus strongest baselineENGINE · ObjectNet B0 Inc20average accuracy: 52.62 -> 59.11Improved
use OpenAI pretrained CLIP weights instead of LAION-400M (ablation)ENGINE · UCF B0 Inc10accuracyNo clear effect
ENGINE vs Finetune CLIP in each stageFinetune · Aircraft B0 Inc10average accuracy: 69.69 -> 3.16Hurt

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