中文
相关论文

相关论文: Prompt Tuning Vision Language Models with Margin R…

200 篇论文

Large pre-trained vision-language models such as CLIP have demonstrated great potential in zero-shot transferability to downstream tasks. However, to attain optimal performance, the manual selection of prompts is necessary to improve…

计算机视觉与模式识别 · 计算机科学 2024-09-17 Thi Minh Anh Pham , An Duc Nguyen , Cephas Svosve , Vasileios Argyriou , Georgios Tzimiropoulos

Prompt tuning (PT) has long been recognized as an effective and efficient paradigm for transferring large pre-trained vision-language models (VLMs) to downstream tasks by learning a tiny set of context vectors. Nevertheless, in this work,…

计算机视觉与模式识别 · 计算机科学 2025-03-17 Shihan Wu , Ji Zhang , Pengpeng Zeng , Lianli Gao , Jingkuan Song , Heng Tao Shen

Vision-language models (VLMs) classify the query video by calculating a similarity score between the visual features and text-based class label representations. Recently, large language models (LLMs) have been used to enrich the text-based…

计算机视觉与模式识别 · 计算机科学 2023-10-25 Adeel Yousaf , Muzammal Naseer , Salman Khan , Fahad Shahbaz Khan , Mubarak Shah

Few-shot learning (FSL) has attracted increasing attention in recent years but remains challenging, due to the intrinsic difficulty in learning to generalize from a few examples. This paper proposes an adaptive margin principle to improve…

计算机视觉与模式识别 · 计算机科学 2020-05-29 Aoxue Li , Weiran Huang , Xu Lan , Jiashi Feng , Zhenguo Li , Liwei Wang

Pre-trained language models have shown excellent results in few-shot learning scenarios using in-context learning. Although it is impressive, the size of language models can be prohibitive to make them usable in on-device applications, such…

计算与语言 · 计算机科学 2022-04-27 Navid Rezaei , Marek Z. Reformat

We investigate the generalization capabilities of small language models under two popular adaptation paradigms: few-shot prompting and supervised fine-tuning. While prompting is often favored for its parameter efficiency and flexibility, it…

人工智能 · 计算机科学 2025-06-26 Rahul Raja , Arpita Vats

Large Language Models (LLMs) have recently demonstrated impressive few-shot learning capabilities through in-context learning (ICL). However, ICL performance is highly dependent on the choice of few-shot demonstrations, making the selection…

计算与语言 · 计算机科学 2025-06-03 Soumya Suvra Ghosal , Soumyabrata Pal , Koyel Mukherjee , Dinesh Manocha

Pre-trained large language models can efficiently interpolate human-written prompts in a natural way. Multitask prompted learning can help generalization through a diverse set of tasks at once, thus enhancing the potential for more…

计算与语言 · 计算机科学 2022-12-22 M Saiful Bari , Aston Zhang , Shuai Zheng , Xingjian Shi , Yi Zhu , Shafiq Joty , Mu Li

Visual Prompt Tuning (VPT) of pre-trained Vision Transformers (ViTs) has proven highly effective as a parameter-efficient fine-tuning technique for adapting large models to downstream tasks with limited data. Its parameter efficiency makes…

计算机视觉与模式识别 · 计算机科学 2026-05-12 M Yashwanth , Sharannya Ghosh , Aditay Tripathi , Anirban Chakraborty

Understanding the vulnerability of large-scale pre-trained vision-language models like CLIP against adversarial attacks is key to ensuring zero-shot generalization capacity on various downstream tasks. State-of-the-art defense mechanisms…

计算机视觉与模式识别 · 计算机科学 2024-05-21 Fan Yang , Mingxuan Xia , Sangzhou Xia , Chicheng Ma , Hui Hui

Tuning pre-trained language models (PLMs) with task-specific prompts has been a promising approach for text classification. Particularly, previous studies suggest that prompt-tuning has remarkable superiority in the low-data scenario over…

计算与语言 · 计算机科学 2022-03-21 Shengding Hu , Ning Ding , Huadong Wang , Zhiyuan Liu , Jingang Wang , Juanzi Li , Wei Wu , Maosong Sun

Prompt tuning is a parameter-efficient approach to adapting pre-trained language models to downstream tasks. Although prompt tuning has been shown to match the performance of full model tuning when training data is sufficient, it tends to…

计算与语言 · 计算机科学 2023-05-09 Tianxiang Sun , Zhengfu He , Qin Zhu , Xipeng Qiu , Xuanjing Huang

Large pre-trained language models (PLMs) are at the forefront of advances in Natural Language Processing. One widespread use case of PLMs is "prompting" - or in-context learning - where a user provides a description of a task and some…

计算与语言 · 计算机科学 2024-03-12 Christopher Toukmaji

Vision-language models (VLMs) like CLIP have been cherished for their ability to perform zero-shot visual recognition on open-vocabulary concepts. This is achieved by selecting the object category whose textual representation bears the…

计算机视觉与模式识别 · 计算机科学 2024-12-06 Shaunak Halbe , Junjiao Tian , K J Joseph , James Seale Smith , Katherine Stevo , Vineeth N Balasubramanian , Zsolt Kira

Although soft prompt tuning is effective in efficiently adapting Vision-Language (V&L) models for downstream tasks, it shows limitations in dealing with distribution shifts. We address this issue with Attribute-Guided Prompt Tuning (ArGue),…

计算机视觉与模式识别 · 计算机科学 2024-03-14 Xinyu Tian , Shu Zou , Zhaoyuan Yang , Jing Zhang

Effectively finetuning pretrained language models (PLMs) is critical for their success in downstream tasks. However, PLMs may have risks in overfitting the pretraining tasks and data, which usually have gap with the target downstream tasks.…

计算与语言 · 计算机科学 2022-03-24 Chuhan Wu , Fangzhao Wu , Tao Qi , Yongfeng Huang , Xing Xie

Prompt-based methods have been successfully applied in sentence-level few-shot learning tasks, mostly owing to the sophisticated design of templates and label words. However, when applied to token-level labeling tasks such as NER, it would…

计算与语言 · 计算机科学 2022-11-24 Ruotian Ma , Xin Zhou , Tao Gui , Yiding Tan , Linyang Li , Qi Zhang , Xuanjing Huang

The zero-shot performance of existing vision-language models (VLMs) such as CLIP is limited by the availability of large-scale, aligned image and text datasets in specific domains. In this work, we leverage two complementary sources of…

计算机视觉与模式识别 · 计算机科学 2024-04-05 Oindrila Saha , Grant Van Horn , Subhransu Maji

Integrating image and text data through multi-modal learning has emerged as a new approach in medical imaging research, following its successful deployment in computer vision. While considerable efforts have been dedicated to establishing…

计算机视觉与模式识别 · 计算机科学 2024-09-09 Fereshteh Shakeri , Yunshi Huang , Julio Silva-Rodríguez , Houda Bahig , An Tang , Jose Dolz , Ismail Ben Ayed

This paper surveys and organizes research works in a new paradigm in natural language processing, which we dub "prompt-based learning". Unlike traditional supervised learning, which trains a model to take in an input x and predict an output…

计算与语言 · 计算机科学 2021-07-30 Pengfei Liu , Weizhe Yuan , Jinlan Fu , Zhengbao Jiang , Hiroaki Hayashi , Graham Neubig