paper-with-me

홈 › Papers

DLIP: Distilling Language-Image Pre-training

2023-08-24 · Huafeng Kuang, Jie Wu, Xiawu Zheng, Ming Li, Xuefeng Xiao, Rui Wang, Min Zheng, Rongrong Ji

Vision-Language Pre-training (VLP) shows remarkable progress with the assistance of extremely heavy parameters, which challenges deployment in real applications. Knowledge distillation is well recognized as the essential procedure in model compression. However, existing knowledge distillation techniques lack an in-depth investigation and analysis of VLP, and practical guidelines for VLP-oriented distillation are still not yet explored. In this paper, we present DLIP, a simple yet efficient Distilling Language-Image Pre-training framework, through which we investigate how to distill a light VLP model. Specifically, we dissect the model distillation from multiple dimensions, such as the architecture characteristics of different modules and the information transfer of different modalities. We conduct comprehensive experiments and provide insights on distilling a light but performant VLP model. Experimental results reveal that DLIP can achieve a state-of-the-art accuracy/efficiency trade-off across diverse cross-modal tasks, e.g., image-text retrieval, image captioning and visual question answering. For example, DLIP compresses BLIP by 1.9x, from 213M to 108M parameters, while achieving comparable or better performance. Furthermore, DLIP succeeds in retaining more than 95% of the performance with 22.4% parameters and 24.8% FLOPs compared to the teacher model and accelerates inference speed by 2.7x.

📄 PDF Abstract BibTeX arXiv:2308.12956

Code (0)

등록된 구현이 없습니다.

Tasks

Image CaptioningImage-text RetrievalKnowledge DistillationModel CompressionQuestion AnsweringRetrievalText RetrievalVisual Question Answering

Methods 이 논문이 사용한 방법론

BLIP Vision-Language Pre-training (VLP) has advanced the performance for many vision-language tasks. However, most existing pre-trained models only excel in either understanding-based…
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.…
SPEED The monocular depth estimation (MDE) is the task of estimating depth from a single frame. This information is an essential knowledge in many computer vision tasks such as scene…

Similar Papers 제목 키워드 기반

DLiPath: A Benchmark for the Comprehensive Assessment of Donor Liver Based on Histopathological Image Dataset

2025-05-30 · Liangrui Pan, Xingchen Li, Zhongyi Chen, Ling Chu 외

Pathologists comprehensive evaluation of donor liver biopsies provides crucial information for accepting or discarding potential grafts. However, rapidly and accurately obtaining these assessments intraoperatively poses …

Multiple Instance Learningwhole slide images

Cross-modal Audio-visual Co-learning for Text-independent Speaker Verification

2023-02-22 · Meng Liu, Kong Aik Lee, Longbiao Wang, Hanyi Zhang 외

Visual speech (i.e., lip motion) is highly related to auditory speech due to the co-occurrence and synchronization in speech production. This paper investigates this correlation and proposes a cross-modal speech co-learn…

Speaker VerificationText-Independent Speaker Verification

KD-DLGAN: Data Limited Image Generation via Knowledge Distillation

2023-03-30 · CVPR 2023 1 · Kaiwen Cui, Yingchen Yu, Fangneng Zhan, Shengcai Liao 외

Generative Adversarial Networks (GANs) rely heavily on large-scale training data for training high-quality image generation models. With limited training data, the GAN discriminator often suffers from severe overfitting …

DiversityImage GenerationKnowledge Distillation

Moonshine: Distilling Game Content Generators into Steerable Generative Models

2024-08-18 · Yuhe Nie, Michael Middleton, Tim Merino, Nidhushan Kanagaraja 외

Procedural Content Generation via Machine Learning (PCGML) has enhanced game content creation, yet challenges in controllability and limited training data persist. This study addresses these issues by distilling a constr…

Language ModelingLanguage ModellingLarge Language Model

Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes

2023-05-03 · Cheng-Yu Hsieh, Chun-Liang Li, Chih-Kuan Yeh, Hootan Nakhost 외

Deploying large language models (LLMs) is challenging because they are memory inefficient and compute-intensive for practical applications. In reaction, researchers train smaller task-specific models by either finetuning…