paper-with-me

Papers

Adapter-state Sharing CLIP for Parameter-efficient Multimodal Sarcasm Detection

2025-07-06 · Soumyadeep Jana, Sahil Danayak, Sanasam Ranbir Singh arxiv

The growing prevalence of multimodal image-text sarcasm on social media poses challenges for opinion mining systems. Existing approaches rely on full fine-tuning of large models, making them unsuitable to adapt under resource-constrained settings. While recent parameter-efficient fine-tuning (PEFT) methods offer promise, their off-the-shelf use underperforms on complex tasks like sarcasm detection. We propose AdS-CLIP (Adapter-state Sharing in CLIP), a lightweight framework built on CLIP that inserts adapters only in the upper layers to preserve low-level unimodal representations in the lower layers and introduces a novel adapter-state sharing mechanism, where textual adapters guide visual ones to promote efficient cross-modal learning in the upper layers. Experiments on two public benchmarks demonstrate that AdS-CLIP not only outperforms standard PEFT methods but also existing multimodal baselines with significantly fewer trainable parameters.

📄 PDF Abstract BibTeX arXiv:2507.04508

Code (0)

등록된 구현이 없습니다.

Tasks

parameter-efficient fine-tuningSarcasm DetectionOpinion Mining

Similar Papers 제목 키워드 기반

MV-Adapter: Multimodal Video Transfer Learning for Video Text Retrieval

2023-01-19 · CVPR 2024 1 · Xiaojie Jin, BoWen Zhang, Weibo Gong, Kai Xu 외

State-of-the-art video-text retrieval (VTR) methods typically involve fully fine-tuning a pre-trained model (e.g. CLIP) on specific datasets. However, this can result in significant storage costs in practical application…

RetrievalText RetrievalTransfer LearningVideo-Text Retrieval

RMAdapter: Reconstruction-based Multi-Modal Adapter for Vision-Language Models

2025-12-07 · Xiang Lin, Weixin Li, Shu Guo, Lihong Wang 외 arxiv

Pre-trained Vision-Language Models (VLMs), \textit{e.g.} CLIP, have become essential tools in multimodal transfer learning. However, fine-tuning VLMs in few-shot scenarios poses significant challenges in balancing task-s…

parameter-efficient fine-tuningDomain GeneralizationGeneral KnowledgeTransfer Learning

MA-CLIP: Towards Modality-Agnostic Contrastive Language-Image Pre-training

2021-09-29 · Haoxuan You, Luowei Zhou, Bin Xiao, Noel C Codella 외

Large-scale multimodal contrastive pretraining has demonstrated great utility to support high performance in a range of downstream tasks by mapping multiple modalities into a shared embedding space. Typically, this has e…

CapS-Adapter: Caption-based MultiModal Adapter in Zero-Shot Classification

2024-05-26 · Qijie Wang, Guandu Liu, Bin Wang

Recent advances in vision-language foundational models, such as CLIP, have demonstrated significant strides in zero-shot classification. However, the extensive parameterization of models like CLIP necessitates a resource…

zero-shot-classificationZero-Shot Learning

MIP: CLIP-based Image Reconstruction from PEFT Gradients

2024-02-26 · Peiheng Zhou, Ming Hu, Xiaofei Xie, Yihao Huang 외

Contrastive Language-Image Pre-training (CLIP) model, as an effective pre-trained multimodal neural network, has been widely used in distributed machine learning tasks, especially Federated Learning (FL). Typically, CLIP…

Federated LearningImage Reconstructionparameter-efficient fine-tuningReconstruction Attack