Open-source research

IISAN-Versa

A decoupled parameter-efficient adaptation framework for symmetric and asymmetric multimodal foundation models in recommendation.

Year
2025
Status
active
Focus
Python · PEFT · Sequential Recommendation

What it does

IISAN-Versa adapts multimodal foundation models for sequential recommendation while reducing the memory and training costs associated with full fine-tuning.

Its decoupled side-adapter design supports both symmetric and asymmetric text-image encoder combinations. The project also covers alignment between encoders with different depths and representation dimensions.

Evidence and scope

This page is a concise guide to the work. Use the primary paper for the experimental setup, quantitative results, limitations, and formal claims; use the repository for the implementation and current reproduction instructions.