Adapting Dual-encoder Vision-language Models for Paraphrased Retrieval

• URL: http://arxiv.org/abs/2405.03190v1
• Date: Mon, 6 May 2024 06:30:17 GMT
• Title: Adapting Dual-encoder Vision-language Models for Paraphrased Retrieval
• Authors: Jiacheng Cheng, Hijung Valentina Shin, Nuno Vasconcelos, Bryan Russell, Fabian Caba Heilbron,
• Abstract summary: We consider the task of paraphrased text-to-image retrieval where a model aims to return similar results given a pair of paraphrased queries. We train a dual-encoder model starting from a language model pretrained on a large text corpus. Compared to public dual-encoder models such as CLIP and OpenCLIP, the model trained with our best adaptation strategy achieves a significantly higher ranking similarity for paraphrased queries.
• Score: 55.90407811819347
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In the recent years, the dual-encoder vision-language models (\eg CLIP) have achieved remarkable text-to-image retrieval performance. However, we discover that these models usually results in very different retrievals for a pair of paraphrased queries. Such behavior might render the retrieval system less predictable and lead to user frustration. In this work, we consider the task of paraphrased text-to-image retrieval where a model aims to return similar results given a pair of paraphrased queries. To start with, we collect a dataset of paraphrased image descriptions to facilitate quantitative evaluation for this task. We then hypothesize that the undesired behavior of existing dual-encoder model is due to their text towers which are trained on image-sentence pairs and lack the ability to capture the semantic similarity between paraphrased queries. To improve on this, we investigate multiple strategies for training a dual-encoder model starting from a language model pretrained on a large text corpus. Compared to public dual-encoder models such as CLIP and OpenCLIP, the model trained with our best adaptation strategy achieves a significantly higher ranking similarity for paraphrased queries while maintaining similar zero-shot classification and retrieval accuracy.

Related papers

• Understanding Retrieval Robustness for Retrieval-Augmented Image Captioning [21.04172981071809]
  We analyze the robustness of a retrieval-augmented captioning model SmallCap. We propose to train the model by sampling retrieved captions from more diverse sets.
  arXiv  Detail & Related papers  (2024-06-04T12:41:54Z)
• Fine-tuning CLIP Text Encoders with Two-step Paraphrasing [83.3736789315201]
  We introduce a straightforward fine-tuning approach to enhance the representations of CLIP models for paraphrases. Our model, which we call ParaCLIP, exhibits significant improvements over baseline CLIP models across various tasks.
  arXiv  Detail & Related papers  (2024-02-23T06:11:50Z)
• Exposing and Mitigating Spurious Correlations for Cross-Modal Retrieval [89.30660533051514]
  Cross-modal retrieval methods are the preferred tool to search databases for the text that best matches a query image and vice versa. Image-text retrieval models commonly learn to spurious correlations in the training data, such as frequent object co-occurrence. We introduce ODmAP@k, an object decorrelation metric that measures a model's robustness to spurious correlations in the training data.
  arXiv  Detail & Related papers  (2023-04-06T21:45:46Z)
• What Are You Token About? Dense Retrieval as Distributions Over the Vocabulary [68.77983831618685]
  We propose to interpret the vector representations produced by dual encoders by projecting them into the model's vocabulary space. We show that the resulting projections contain rich semantic information, and draw connection between them and sparse retrieval.
  arXiv  Detail & Related papers  (2022-12-20T16:03:25Z)
• UnifieR: A Unified Retriever for Large-Scale Retrieval [84.61239936314597]
  Large-scale retrieval is to recall relevant documents from a huge collection given a query. Recent retrieval methods based on pre-trained language models (PLM) can be coarsely categorized into either dense-vector or lexicon-based paradigms. We propose a new learning framework, UnifieR which unifies dense-vector and lexicon-based retrieval in one model with a dual-representing capability.
  arXiv  Detail & Related papers  (2022-05-23T11:01:59Z)
• Zero-Shot Image-to-Text Generation for Visual-Semantic Arithmetic [72.60554897161948]
  Recent text-to-image matching models apply contrastive learning to large corpora of uncurated pairs of images and sentences. In this work, we repurpose such models to generate a descriptive text given an image at inference time. The resulting captions are much less restrictive than those obtained by supervised captioning methods.
  arXiv  Detail & Related papers  (2021-11-29T11:01:49Z)
• Evaluating Contrastive Models for Instance-based Image Retrieval [6.393147386784114]
  We evaluate contrastive models for the task of image retrieval. We find that models trained using contrastive methods perform on-par with (and outperforms) a pre-trained baseline trained on the ImageNet labels.
  arXiv  Detail & Related papers  (2021-04-30T12:05:23Z)
• A Multi-cascaded Model with Data Augmentation for Enhanced Paraphrase Detection in Short Texts [1.6758573326215689]
  We present a data augmentation strategy and a multi-cascaded model for improved paraphrase detection in short texts. Our model is both wide and deep and provides greater robustness across clean and noisy short texts.
  arXiv  Detail & Related papers  (2019-12-27T12:10:10Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.