MEGA Hub

SE-MoLoRA: Shared-Expert LoRA Adapters for Domain-Specific Photographic Assessment

Authors

Do you know Bishwash Khanal?You can claim authorship or link another user.Do you know Anlan Zhang?You can claim authorship or link another user.Do you know Sasu Tarkoma?You can claim authorship or link another user.Do you know Tommi Mikkonen?You can claim authorship or link another user.Do you know Abhishek Kumar?You can claim authorship or link another user.

Abstract

Vision-language models can describe images fluently, but they often fail to provide actionable photographic critique because semantic content and aesthetic judgment remain entangled. We propose SE-MoLoRA, a modular parameter-efficient adaptation framework for domain-specific photographic assessment. The method separates general photographic knowledge from specialist residual judgments using an always-active shared LoRA expert and routed adapters for composition, lighting, and technical quality. A lightweight query router selects the relevant specialist, enabling targeted critique without training separate full models. A rank-64 shared adapter captures broad photographic vocabulary, while rank-32 specialists learn domain-specific residuals with an orthogonal regularization penalty that encourages disentangled representations. Training data is obtained by distilling the Reddit Photo Critique Dataset into domain-labeled critique samples. On held-out critique generation, SE-MoLoRA improves BERTScore-F1 from 0.2317 to 0.4215 over monolithic LoRA and is preferred in 84.6\% of pairwise comparisons, while using fewer active parameters than separate specialist models. SVD-based ablation study shows that shared-specialist decomposition and orthogonal regularization reduce expert overlap. These results demonstrate that modular adaptation improves controllability and specificity in multimodal photographic critique.

Community

00