paper-with-me

홈 › Papers

Enabling Multimodal Generation on CLIP via Vision-Language Knowledge Distillation

2021-11-16 · ACL ARR November 2021 11 · Anonymous

The recent large-scale vision-language pre-training (VLP) of dual-stream architectures (e.g., CLIP) with a tremendous amount of image-text pair data, has shown its superiority on various multimodal alignment tasks. Despite its success, the resulting models are not capable of generative multimodal tasks due to the weak text encoder. To tackle this problem, we propose to augment the dual-stream VLP model with a textual pre-trained language model (PLM) via vision-language knowledge distillation (VLKD), enabling the capability for multimodal generation. VLKD is pretty data- and computation-efficient compared to the pre-training from scratch. Experimental results show that the resulting model has strong zero-shot performance on multimodal generation tasks, such as open-ended visual question answering and image captioning. For example, it achieves 39.7% zero-shot accuracy on the VQA 2.0 dataset, surpassing the previous state-of-the-art zero-shot model with 14x fewer parameters. Furthermore, the original text processing ability of the PLM is maintained after VLKD, which makes our model versatile for both multimodal and unimodal tasks.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

Image CaptioningKnowledge DistillationLanguage ModelingLanguage Modellingmultimodal generationQuestion AnsweringVisual Question AnsweringVisual Question Answering (VQA)

Methods 이 논문이 사용한 방법론

Knowledge Distillation A very simple way to improve the performance of almost any machine learning algorithm is to train many different models on the same data and then to average their predictions.…

Similar Papers 제목 키워드 기반

Enabling Multimodal Generation on CLIP via Vision-Language Knowledge Distillation

2022-03-12 · Findings (ACL) 2022 5 · Wenliang Dai, Lu Hou, Lifeng Shang, Xin Jiang 외

The recent large-scale vision-language pre-training (VLP) of dual-stream architectures (e.g., CLIP) with a tremendous amount of image-text pair data, has shown its superiority on various multimodal alignment tasks. Despi…

Image CaptioningKnowledge DistillationLanguage ModelingLanguage Modelling+4

Multi-Faceted Multimodal Monosemanticity

2025-02-16 · Hanqi Yan, Xiangxiang Cui, Lu Yin, Paul Pu Liang 외

Humans experience the world through multiple modalities, such as, vision, language, and speech, making it natural to explore the commonality and distinctions among them. In this work, we take a data-driven approach to ad…

AttributeImage GenerationText to Image GenerationText-to-Image Generation

MMLGNet: Cross-Modal Alignment of Remote Sensing Data using CLIP

2026-01-13 · Aditya Chaudhary, Sneha Barman, Mainak Singha, Ankit Jha 외 arxiv

In this paper, we propose a novel multimodal framework, Multimodal Language-Guided Network (MMLGNet), to align heterogeneous remote sensing modalities like Hyperspectral Imaging (HSI) and LiDAR with natural language sema…

Contrastive Learning

Multimodal Abstractive Summarization of Instructional Videos with Vision-Language Models

2026-05-12 · Maham Nazir, Muhammad Aqeel, Richong Zhang, Francesco Setti arxiv

Multimodal video summarization requires visual features that align semantically with language generation. Traditional approaches rely on CNN features trained for object classification, which represent visual concepts as …

Video Summarization

VL-CLIP: Enhancing Multimodal Recommendations via Visual Grounding and LLM-Augmented CLIP Embeddings

2025-07-22 · Ramin Giahi, Kehui Yao, Sriram Kollipara, Kai Zhao 외 arxiv

Multimodal learning plays a critical role in e-commerce recommendation platforms today, enabling accurate recommendations and product understanding. However, existing vision-language models, such as CLIP, face key challe…

Recommendation SystemsVisual Grounding