AI Paper Atlas

Research map / Continual learning / Continual fine-tuning of LLMs

Achieving More with Less: Additive Prompt Tuning for Rehearsal-Free Class-Incremental Learning

Haoran Chen, Ping Wang, Zihan Zhou, Xu Zhang, Zuxuan Wu, Yu-Gang Jiang

ICCV 2025 · Continual learning › Continual fine-tuning of LLMs · approach: Synthetic data and merging for CLIP · 6 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 progressive prompt fusion (PPF) at inference (ablation)APT · Split CIFAR-100 (10 tasks)Average accuracy: 88.17 -> 88.88Improved
add PPF at inference (ablation)APT · Split ImageNet-R (10 tasks)Average accuracy: 78.56 -> 79.40Improved
add PPF at inference (ablation)APT · Split CIFAR-100 (10 tasks)Forgetting: 7.05 -> 3.47Improved
add PPF at inference (ablation)APT · Split ImageNet-R (10 tasks)Forgetting: 8.83 -> 4.38Improved
add prompts to CLS key/value instead of CLS input (ablation)APT · Split CIFAR-100 (10 tasks)Average accuracy: 87.93 -> 88.88Improved
add prompts to CLS key/value instead of CLS input (ablation)APT · Split ImageNet-R (10 tasks)Average accuracy: 78.31 -> 79.40Improved
add prompts to CLS key/value instead of CLS input (ablation)APT · Split CIFAR-100 (10 tasks)Forgetting: 5.24 -> 3.47Improved
add prompts to CLS key/value instead of CLS input (ablation)APT · Split ImageNet-R (10 tasks)Forgetting: 5.18 -> 4.38No clear effect
vary PPF coefficient alpha from 0.2 to 0.8 (ablation)APT · Split CIFAR-100 and ImageNet-RAverage accuracyNo clear effect
extend sequence from 10 to 20 tasksAPT · Split CIFAR-100Average accuracy: 88.88 -> 86.37Hurt
APT vs CPrompt on 20-task ImageNet-R forgettingAPT · Split ImageNet-R (20 tasks)Forgetting: 6.41 -> 6.91Hurt
replace concatenated prompts with additive CLS prompts vs VPT-deepAPT · FGVC+VTAB averageAverage accuracy: 84.4 -> 86.3Improved

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