paper-with-me

홈 › Papers

Prompt Reinjection: Alleviating Prompt Forgetting in Multimodal Diffusion Transformers

2026-02-06 · Yuxuan Yao, Yuxuan Chen, Hui Li, Kaihui Cheng, Qipeng Guo, Yuwei Sun, Zilong Dong, Jingdong Wang, Siyu Zhu arxiv

Multimodal Diffusion Transformers (MMDiTs) for text-to-image generation maintain separate text and image branches, with bidirectional information flow between text tokens and visual latents throughout denoising. In this setting, we observe a prompt forgetting phenomenon: the semantics of the prompt representation in the text branch is progressively forgotten as depth increases. We further verify this effect on three representative MMDiTs--SD3, SD3.5, and FLUX.1 by probing linguistic attributes of the representations over the layers in the text branch. Motivated by these findings, we introduce a training-free approach, prompt reinjection, which reinjects prompt representations from early layers into later layers to alleviate this forgetting. Experiments on GenEval, DPG, and T2I-CompBench++ show consistent gains in instruction-following capability, along with improvements on metrics capturing preference, aesthetics, and overall text--image generation quality.

📄 PDF Abstract BibTeX arXiv:2602.06886

Code (0)

등록된 구현이 없습니다.

Tasks

Text-to-Image Generation

Similar Papers 제목 키워드 기반

ModalPrompt:Dual-Modality Guided Prompt for Continual Learning of Large Multimodal Models

2024-10-08 · Fanhu Zeng, Fei Zhu, Haiyang Guo, Xu-Yao Zhang 외

Large Multimodal Models (LMMs) exhibit remarkable multi-tasking ability by learning mixed datasets jointly. However, novel tasks would be encountered sequentially in dynamic world, and continually fine-tuning LMMs often …

Continual LearningPrompt LearningTransfer Learning

Making Pre-trained Language Models Better Continual Few-Shot Relation Extractors

2024-02-24 · Shengkun Ma, Jiale Han, Yi Liang, Bo Cheng

Continual Few-shot Relation Extraction (CFRE) is a practical problem that requires the model to continuously learn novel relations while avoiding forgetting old ones with few labeled training data. The primary challenges…

Contrastive LearningPrompt LearningRelationRelation Extraction

FPPL: An Efficient and Non-IID Robust Federated Continual Learning Framework

2024-11-04 · Yuchen He, Chuyun Shen, Xiangfeng Wang, Bo Jin

Federated continual learning (FCL) aims to learn from sequential data stream in the decentralized federated learning setting, while simultaneously mitigating the catastrophic forgetting issue in classical continual learn…

Continual LearningContrastive LearningFederated LearningPrompt Learning

Beyond Anti-Forgetting: Multimodal Continual Instruction Tuning with Positive Forward Transfer

2024-01-17 · Junhao Zheng, Qianli Ma, Zhen Liu, Binquan Wu 외

Multimodal Continual Instruction Tuning (MCIT) enables Multimodal Large Language Models (MLLMs) to meet continuously emerging requirements without expensive retraining. MCIT faces two major obstacles: catastrophic forget…

Multimodal Prompting with Missing Modalities for Visual Recognition

2023-03-06 · CVPR 2023 1 · Yi-Lun Lee, Yi-Hsuan Tsai, Wei-Chen Chiu, Chen-Yu Lee

In this paper, we tackle two challenges in multimodal learning for visual recognition: 1) when missing-modality occurs either during training or testing in real-world situations; and 2) when the computation resources are…

Prompt Learning