AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

Enhanced Continual Learning of Vision-Language Models with Model Fusion

Haoyuan Gao, Zicong Zhang, Yuqi Wei, Linglan Zhao, Guilin Li, Yexin Li, Bo Wang, Linghe Kong, Weiran Huang

ICLR 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Adapting VLMs for zero-shot classification · 2 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 rescalers to the task triggers (ablation)ConDU (FT) · MTIL (order I)Average accuracy (%): 74.2 -> 78.8Improved
add rescalers to the task triggers (ablation)ConDU (FT) · MTIL (order I)Last accuracy (%): 77.1 -> 87.1Improved
add rescalers to the task triggers (ablation)ConDU (FT) · MTIL (order I)Transfer accuracy (%): 70.4 -> 70.8Improved
compute prototype similarity with pre-trained model features instead of task-expert features (ablation)ConDU (FT) · MTIL (order I)Transfer accuracy (%): 67.7 -> 70.8Improved
compute prototype similarity with pre-trained model features instead of task-expert features (ablation)ConDU (FT) · MTIL (order I)Average accuracy (%): 77.8 -> 78.8Improved
aggregate K=5 task experts instead of K=1 (ablation)ConDU (FT) · MTIL (order I)Transfer accuracy (%): 69.5 -> 70.8Improved
increase K from 5 to 11 (ablation)ConDU (FT) · MTIL (order I)Transfer accuracy (%): 70.8 -> 70.2Hurt
aggregate decoupled task experts instead of using the unified model directly (ablation)ConDU (FT) · MTIL (order I)Transfer accuracy (%): 47.1 -> 70.8Improved
run on RTX 4090 instead of Ascend 910BConDU (FT) · MTIL (order I)Transfer accuracy (%): 70.2 -> 70.8No clear effect
use LoRA instead of full fine-tuningConDU · Few-shot MTILAverage accuracy (%): 72.3 -> 72.7Improved
use LoRA instead of full fine-tuningConDU · TinyImageNet class-incremental (10 steps)accuracy (%): 71.74 -> 68.80Hurt
use ConDU instead of the best baseline (DPeCLIP)ConDU (FT) · Task-agnostic MTILAverage accuracy (%): 76.1 -> 78.1Improved

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