Extract Free Dense Misalignment from CLIP
- URL: http://arxiv.org/abs/2412.18404v1
- Date: Tue, 24 Dec 2024 12:51:05 GMT
- Title: Extract Free Dense Misalignment from CLIP
- Authors: JeongYeon Nam, Jinbae Im, Wonjae Kim, Taeho Kil,
- Abstract summary: This work proposes a novel approach, dubbed CLIP4DM, for detecting dense misalignments from pre-trained CLIP.
We revamp the gradient-based attribution computation method, enabling negative gradient of individual text tokens to indicate misalignment.
Our method demonstrates state-of-the-art performance among zero-shot models and competitive performance with fine-tuned models.
- Score: 7.0247398611254175
- License:
- Abstract: Recent vision-language foundation models still frequently produce outputs misaligned with their inputs, evidenced by object hallucination in captioning and prompt misalignment in the text-to-image generation model. Recent studies have explored methods for identifying misaligned elements, aiming not only to enhance interpretability but also to improve model performance. However, current approaches primarily rely on large foundation models in a zero-shot manner or fine-tuned models with human annotations, which limits scalability due to significant computational costs. This work proposes a novel approach, dubbed CLIP4DM, for detecting dense misalignments from pre-trained CLIP, specifically focusing on pinpointing misaligned words between image and text. We carefully revamp the gradient-based attribution computation method, enabling negative gradient of individual text tokens to indicate misalignment. We also propose F-CLIPScore, which aggregates misaligned attributions with a global alignment score. We evaluate our method on various dense misalignment detection benchmarks, covering various image and text domains and misalignment types. Our method demonstrates state-of-the-art performance among zero-shot models and competitive performance with fine-tuned models while maintaining superior efficiency. Our qualitative examples show that our method has a unique strength to detect entity-level objects, intangible objects, and attributes that can not be easily detected for existing works. We conduct ablation studies and analyses to highlight the strengths and limitations of our approach. Our code is publicly available at https://github.com/naver-ai/CLIP4DM.
Related papers
- Removing Distributional Discrepancies in Captions Improves Image-Text Alignment [76.31530836622694]
We introduce a model designed to improve the prediction of image-text alignment.
Our approach focuses on generating high-quality training datasets for the alignment task.
We also demonstrate the applicability of our model by ranking the images generated by text-to-image models based on text alignment.
arXiv Detail & Related papers (2024-10-01T17:50:17Z) - DECIDER: Leveraging Foundation Model Priors for Improved Model Failure Detection and Explanation [18.77296551727931]
We propose DECIDER, a novel approach that leverages priors from large language models (LLMs) and vision-language models (VLMs) to detect failures in image models.
DECIDER consistently achieves state-of-the-art failure detection performance, significantly outperforming baselines in terms of the overall Matthews correlation coefficient.
arXiv Detail & Related papers (2024-08-01T07:08:11Z) - Information Theoretic Text-to-Image Alignment [49.396917351264655]
Mutual Information (MI) is used to guide model alignment.
Our method uses self-supervised fine-tuning and relies on a point-wise (MI) estimation between prompts and images.
Our analysis indicates that our method is superior to the state-of-the-art, yet it only requires the pre-trained denoising network of the T2I model itself to estimate MI.
arXiv Detail & Related papers (2024-05-31T12:20:02Z) - Improving Input-label Mapping with Demonstration Replay for In-context
Learning [67.57288926736923]
In-context learning (ICL) is an emerging capability of large autoregressive language models.
We propose a novel ICL method called Sliding Causal Attention (RdSca)
We show that our method significantly improves the input-label mapping in ICL demonstrations.
arXiv Detail & Related papers (2023-10-30T14:29:41Z) - ClipCrop: Conditioned Cropping Driven by Vision-Language Model [90.95403416150724]
We take advantage of vision-language models as a foundation for creating robust and user-intentional cropping algorithms.
We develop a method to perform cropping with a text or image query that reflects the user's intention as guidance.
Our pipeline design allows the model to learn text-conditioned aesthetic cropping with a small dataset.
arXiv Detail & Related papers (2022-11-21T14:27:07Z) - Few-shot Text Classification with Dual Contrastive Consistency [31.141350717029358]
In this paper, we explore how to utilize pre-trained language model to perform few-shot text classification.
We adopt supervised contrastive learning on few labeled data and consistency-regularization on vast unlabeled data.
arXiv Detail & Related papers (2022-09-29T19:26:23Z) - No Token Left Behind: Explainability-Aided Image Classification and
Generation [79.4957965474334]
We present a novel explainability-based approach, which adds a loss term to ensure that CLIP focuses on all relevant semantic parts of the input.
Our method yields an improvement in the recognition rate, without additional training or fine-tuning.
arXiv Detail & Related papers (2022-04-11T07:16:39Z) - Detection and Captioning with Unseen Object Classes [12.894104422808242]
Test images may contain visual objects with no corresponding visual or textual training examples.
We propose a detection-driven approach based on a generalized zero-shot detection model and a template-based sentence generation model.
Our experiments show that the proposed zero-shot detection model obtains state-of-the-art performance on the MS-COCO dataset.
arXiv Detail & Related papers (2021-08-13T10:43:20Z) - Automatic Generation of Contrast Sets from Scene Graphs: Probing the
Compositional Consistency of GQA [16.95631509102115]
supervised models often exploit data artifacts to achieve good test scores while their performance severely degrades on samples outside their training distribution.
We present a novel method which leverages rich semantic input representation to automatically generate contrast sets for the visual question answering task.
We find that, despite GQA's compositionality and carefully balanced label distribution, two high-performing models drop 13-17% in accuracy compared to the original test set.
arXiv Detail & Related papers (2021-03-17T12:19:25Z) - Word Shape Matters: Robust Machine Translation with Visual Embedding [78.96234298075389]
We introduce a new encoding of the input symbols for character-level NLP models.
It encodes the shape of each character through the images depicting the letters when printed.
We name this new strategy visual embedding and it is expected to improve the robustness of NLP models.
arXiv Detail & Related papers (2020-10-20T04:08:03Z)
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.