{"ID":2897970,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.04508","arxiv_id":"2507.04508","title":"Adapter-state Sharing CLIP for Parameter-efficient Multimodal Sarcasm Detection","abstract":"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.","short_abstract":"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 promis...","url_abs":"https://arxiv.org/abs/2507.04508","url_pdf":"https://arxiv.org/pdf/2507.04508v2","authors":"[\"Soumyadeep Jana\",\"Sahil Danayak\",\"Sanasam Ranbir Singh\"]","published":"2025-07-06T18:51:00Z","proceeding":"cs.CL","tasks":"[\"cs.CL\"]","methods":"[]","has_code":false}