Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Few-Shot Domain Incremental Learning via Continual Vision-Language Consolidation

About

Existing domain-incremental learning (DIL) strategies call for massive amounts of data to adapt to new domains and suffer from the overfitting problem in the case of data scarcity. This paper puts forward a relatively uncharted problem, namely, few-shot domain incremental learning (FSDIL), taking into account the problem of extreme data shortages in the realm of DIL. A novel algorithm, namely Continual Vision-Language Consolidation (CVLC), is proposed to address the FSDIL problem, where the key idea lies in the concept of latent space reservation in the base domain coupled with dual coalescent projection (DCP) as a parameter-efficient fine-tuning method. First, the vision prototype is calibrated while multiple templates and synonyms are generated via LLMs to induce the language prototype. The vision and language prototypes are fused. Adaptation to never-ending arrivals of new domains is done by the DCP technique, fine-tuned in such a way to prepare the model to unseen domains via latent-space reservations committed in the base domain. CVLC is structured under shared and domain-specific components to combine general knowledge and domain-specific details. The advantage of our approach is demonstrated through a range of benchmark problems and comparisons with prior arts, in which CVLC outperforms them by up to a 16% gap. Our codes are shared publicly in https://github.com/Naeem-Paeedeh/CVLC .

Naeem Paeedeh, Mahardhika Pratama, Wolfgang Mayer, Mukesh Prasad, Weiping Ding, Yew-Soon Ong• 2026

Related benchmarks

TaskDatasetResultRank
Domain-incremental learningCORe50
Avg Accuracy (A)96.96
108
Few-Shot Domain Incremental LearningDomainNet
AA*75.11
43
Domain-incremental learningCDDB-Hard 1-shot
AA*87.32
11
Domain-incremental learningCDDB-Hard 2-shot
Average Accuracy (AA*)87.77
11
Domain-incremental learningCDDB-Hard 4-shot
AA*88.64
11
Domain-incremental learningCDDB-Hard 8-shot
AA*89.29
11
Incremental Image ClassificationCORe50
Average Accuracy96.87
6
Incremental Image ClassificationDomainNet
Average Accuracy75.03
6
Showing 8 of 8 rows

Other info

Follow for update