InstructMoLE: Instruction-Guided Mixture of Low-rank Experts for Multi-Conditional Image Generation

TL;DR

InstructMoLE introduces an instruction-guided routing mechanism for mixture of experts in diffusion models, improving multi-conditional image generation fidelity and coherence.

Contribution

It proposes a global instruction-guided routing strategy and an orthogonality loss to enhance expert diversity and global semantic consistency in image generation.

Findings

Outperforms existing LoRA adapters and MoLE variants on multi-conditional benchmarks.

Reduces artifacts like spatial fragmentation and semantic drift.

Enhances compositional control and fidelity to user instructions.

Abstract

Parameter-Efficient Fine-Tuning of Diffusion Transformers (DiTs) for diverse, multi-conditional tasks often suffers from task interference when using monolithic adapters like LoRA. The Mixture of Low-rank Experts (MoLE) architecture offers a modular solution, but its potential is usually limited by routing policies that operate at a token level. Such local routing can conflict with the global nature of user instructions, leading to artifacts like spatial fragmentation and semantic drift in complex image generation tasks. To address these limitations, we introduce InstructMoLE, a novel framework that employs an Instruction-Guided Mixture of Low-Rank Experts. Instead of per-token routing, InstructMoLE utilizes a global routing signal, Instruction-Guided Routing (IGR), derived from the user's comprehensive instruction. This ensures that a single, coherently chosen expert council is applied uniformly across all input tokens, preserving the global semantics and structural integrity of the generation process. To complement this, we introduce an output-space orthogonality loss, which promotes expert functional diversity and mitigates representational collapse. Extensive experiments demonstrate that InstructMoLE significantly outperforms existing LoRA adapters and MoLE variants across challenging multi-conditional generation benchmarks. Our work presents a robust and generalizable framework for instruction-driven fine-tuning of generative models, enabling superior compositional control and fidelity to user intent.