Pareto LoRA: Mitigating Modality Imbalance in Unified Multimodal Models via Pareto-Optimal Gradient Integration

• URL: http://arxiv.org/abs/2606.17296v1
• Date: Mon, 15 Jun 2026 21:05:57 GMT
• Title: Pareto LoRA: Mitigating Modality Imbalance in Unified Multimodal Models via Pareto-Optimal Gradient Integration
• Abstract summary: Unified multimodal models (UMMs) have recently emerged as a promising paradigm for integrating multimodal understanding and generation within a single autoregressive transformer.<n>In this work, we analyze modality imbalance in LoRA-based fine-tuning of UMMs for interleaved text-image generation.<n>We show that vision modality performance degrades substantially more than text modality performance when compared to unimodal counterparts, and that modality-specific gradients can differ by orders of magnitude across various tasks and layers.
• Score: 17.777032374897242
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Unified multimodal models (UMMs) have recently emerged as a promising paradigm for integrating multimodal understanding and generation within a single autoregressive transformer. However, during multimodal instruction tuning, these models often exhibit pronounced modality imbalance: language gradients dominate optimization, thus leading to lower image generation quality, especially under parameter-efficient fine-tuning such as LoRA. In this work, we systematically analyze modality imbalance in LoRA-based fine-tuning of UMMs for interleaved text-image generation. We show that vision modality performance degrades substantially more than text modality performance when compared to unimodal counterparts, and that modality-specific gradients can differ by orders of magnitude across various tasks and layers. Motivated by this observation, we reformulate the multimodal instruction tuning as a bi-objective optimization problem and propose Pareto LoRA, a Pareto-optimal gradient integration strategy that balances the text and image objectives by modulating the gradient direction and strength. Experiments on the CoMM benchmark with Emu2 demonstrate that Pareto LoRA consistently improves multimodal generation balance, achieving up to 44.9% gains in perceptual image quality over vanilla LoRA while maintaining comparable text performance.

Related papers

• BALM: A Model-Agnostic Framework for Balanced Multimodal Learning under Imbalanced Missing Rates [1.9650181409005112]
  BALM is a model-agnostic plug-in framework to achieve balanced multimodal learning under IMR.<n>We show that BALM consistently enhances robustness and improves performance under diverse missing and imbalance settings.
  arXiv  Detail & Related papers  (2026-03-20T07:47:09Z)
• Plug, Play, and Fortify: A Low-Cost Module for Robust Multimodal Image Understanding Models [6.350443894942629]
  Multimodal Weight Allocation Module (MWAM) is a plug-and-play component that dynamically re-balances the contribution of each branch during training.<n>MWAM delivers consistent performance gains across a wide range of tasks and modality combinations.
  arXiv  Detail & Related papers  (2026-02-26T05:51:41Z)
• OmniVL-Guard: Towards Unified Vision-Language Forgery Detection and Grounding via Balanced RL [63.388513841293616]
  Existing forgery detection methods fail to handle the interleaved text, images, and videos prevalent in real-world misinformation.<n>To bridge this gap, this paper targets to develop a unified framework for omnibus vision-language forgery detection and grounding.<n>We propose textbf OmniVL-Guard, a balanced reinforcement learning framework for omnibus vision-language forgery detection and grounding.
  arXiv  Detail & Related papers  (2026-02-11T09:41:36Z)
• MMaDA-Parallel: Multimodal Large Diffusion Language Models for Thinking-Aware Editing and Generation [86.82285754460491]
  We propose a new benchmark designed to evaluate both text and image output modalities.<n>This performance degradation is strongly correlated with poor alignment between the generated reasoning and the final image.<n>We propose a parallel multimodal diffusion framework, MMaDA-Parallel, that enables continuous, bidirectional interaction between text and images.
  arXiv  Detail & Related papers  (2025-11-12T18:58:21Z)
• Query-Kontext: An Unified Multimodal Model for Image Generation and Editing [53.765351127477224]
  Unified Multimodal Models (UMMs) have demonstrated remarkable performance in text-to-image generation (T2I) and editing (TI2I)<n>We introduce Query-Kontext, a novel approach that bridges the VLM and diffusion model via a multimodal kontext'' composed of semantic cues and coarse-grained image conditions encoded from multimodal inputs.<n> Experiments show that our approach matches strong unified baselines and even outperforms task-specific state-of-the-art methods in several cases.
  arXiv  Detail & Related papers  (2025-09-30T17:59:46Z)
• MENTOR: Efficient Multimodal-Conditioned Tuning for Autoregressive Vision Generation Models [30.494968865008513]
  Recent text-to-image models struggle with precise visual control, balancing multimodal inputs, and requiring extensive training for complex image generation.<n>We propose MENTOR, a novel framework for efficient Multimodal-conditioned Tuning for Autoregressive multimodal image generation.<n>Our method delivers superior image reconstruction fidelity, broad task adaptability, and improved training efficiency compared to diffusion-based methods.
  arXiv  Detail & Related papers  (2025-07-13T10:52:59Z)
• Multi-Modality Driven LoRA for Adverse Condition Depth Estimation [61.525312117638116]
  We propose Multi-Modality Driven LoRA (MMD-LoRA) for Adverse Condition Depth Estimation.<n>It consists of two core components: Prompt Driven Domain Alignment (PDDA) and Visual-Text Consistent Contrastive Learning (VTCCL)<n>It achieves state-of-the-art performance on the nuScenes and Oxford RobotCar datasets.
  arXiv  Detail & Related papers  (2024-12-28T14:23:58Z)
• On-the-fly Modulation for Balanced Multimodal Learning [53.616094855778954]
  Multimodal learning is expected to boost model performance by integrating information from different modalities. The widely-used joint training strategy leads to imbalanced and under-optimized uni-modal representations. We propose On-the-fly Prediction Modulation (OPM) and On-the-fly Gradient Modulation (OGM) strategies to modulate the optimization of each modality.
  arXiv  Detail & Related papers  (2024-10-15T13:15:50Z)
• Auto-regressive Image Synthesis with Integrated Quantization [55.51231796778219]
  This paper presents a versatile framework for conditional image generation. It incorporates the inductive bias of CNNs and powerful sequence modeling of auto-regression. Our method achieves superior diverse image generation performance as compared with the state-of-the-art.
  arXiv  Detail & Related papers  (2022-07-21T22:19:17Z)
• Mitigating Modality Collapse in Multimodal VAEs via Impartial Optimization [7.4262579052708535]
  We argue that this effect is a consequence of conflicting gradients during multimodal VAE training. We show how to detect the sub-graphs in the computational graphs where gradients conflict. We empirically show that our framework significantly improves the reconstruction performance, conditional generation, and coherence of the latent space across modalities.
  arXiv  Detail & Related papers  (2022-06-09T13:29:25Z)
• Balanced Multimodal Learning via On-the-fly Gradient Modulation [10.5602074277814]
  Multimodal learning helps to comprehensively understand the world, by integrating different senses. We propose on-the-fly gradient modulation to adaptively control the optimization of each modality, via monitoring the discrepancy of their contribution towards the learning objective.
  arXiv  Detail & Related papers  (2022-03-29T08:26:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.