How structured are the representations in transformer-based vision encoders? An analysis of multi-object representations in vision-language models

• URL: http://arxiv.org/abs/2406.09067v2
• Date: Tue, 18 Jun 2024 12:27:36 GMT
• Title: How structured are the representations in transformer-based vision encoders? An analysis of multi-object representations in vision-language models
• Authors: Tarun Khajuria, Braian Olmiro Dias, Jaan Aru,
• Abstract summary: An extreme form of such abstract representations is symbols. This work estimates the state of such structured representations in vision encoders. We show the network dynamics that cause failure modes of these models on basic downstream tasks in a multi-object scene.
• Score: 2.048226951354646
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Forming and using symbol-like structured representations for reasoning has been considered essential for generalising over novel inputs. The primary tool that allows generalisation outside training data distribution is the ability to abstract away irrelevant information into a compact form relevant to the task. An extreme form of such abstract representations is symbols. Humans make use of symbols to bind information while abstracting away irrelevant parts to utilise the information consistently and meaningfully. This work estimates the state of such structured representations in vision encoders. Specifically, we evaluate image encoders in large vision-language pre-trained models to address the question of which desirable properties their representations lack by applying the criteria of symbolic structured reasoning described for LLMs to the image models. We test the representation space of image encoders like VIT, BLIP, CLIP, and FLAVA to characterise the distribution of the object representations in these models. In particular, we create decoding tasks using multi-object scenes from the COCO dataset, relating the token space to its input content for various objects in the scene. We use these tasks to characterise the network's token and layer-wise information modelling. Our analysis highlights that the CLS token, used for the downstream task, only focuses on a few objects necessary for the trained downstream task. Still, other individual objects are well-modelled separately by the tokens in the network originating from those objects. We further observed a widespread distribution of scene information. This demonstrates that information is far more entangled in tokens than optimal for representing objects similar to symbols. Given these symbolic properties, we show the network dynamics that cause failure modes of these models on basic downstream tasks in a multi-object scene.

Related papers

• Teaching VLMs to Localize Specific Objects from In-context Examples [56.797110842152]
  Vision-Language Models (VLMs) have shown remarkable capabilities across diverse visual tasks. Current VLMs lack a fundamental cognitive ability: learning to localize objects in a scene by taking into account the context. This work is the first to explore and benchmark personalized few-shot localization for VLMs.
  arXiv  Detail & Related papers  (2024-11-20T13:34:22Z)
• Web-Scale Visual Entity Recognition: An LLM-Driven Data Approach [56.55633052479446]
  Web-scale visual entity recognition presents significant challenges due to the lack of clean, large-scale training data. We propose a novel methodology to curate such a dataset, leveraging a multimodal large language model (LLM) for label verification, metadata generation, and rationale explanation. Experiments demonstrate that models trained on this automatically curated data achieve state-of-the-art performance on web-scale visual entity recognition tasks.
  arXiv  Detail & Related papers  (2024-10-31T06:55:24Z)
• Towards Interpreting Visual Information Processing in Vision-Language Models [24.51408101801313]
  Vision-Language Models (VLMs) are powerful tools for processing and understanding text and images. We study the processing of visual tokens in the language model component of LLaVA, a prominent VLM.
  arXiv  Detail & Related papers  (2024-10-09T17:55:02Z)
• ClawMachine: Fetching Visual Tokens as An Entity for Referring and Grounding [67.63933036920012]
  Existing methods, including proxy encoding and geometry encoding, incorporate additional syntax to encode the object's location. This study presents ClawMachine, offering a new methodology that notates an entity directly using the visual tokens. ClawMachine unifies visual referring and grounding into an auto-regressive format and learns with a decoder-only architecture.
  arXiv  Detail & Related papers  (2024-06-17T08:39:16Z)
• Neural Clustering based Visual Representation Learning [61.72646814537163]
  Clustering is one of the most classic approaches in machine learning and data analysis. We propose feature extraction with clustering (FEC), which views feature extraction as a process of selecting representatives from data. FEC alternates between grouping pixels into individual clusters to abstract representatives and updating the deep features of pixels with current representatives.
  arXiv  Detail & Related papers  (2024-03-26T06:04:50Z)
• Helping Hands: An Object-Aware Ego-Centric Video Recognition Model [60.350851196619296]
  We introduce an object-aware decoder for improving the performance of ego-centric representations on ego-centric videos. We show that the model can act as a drop-in replacement for an ego-awareness video model to improve performance through visual-text grounding.
  arXiv  Detail & Related papers  (2023-08-15T17:58:11Z)
• LLM2Loss: Leveraging Language Models for Explainable Model Diagnostics [5.33024001730262]
  We propose an approach that can provide semantic insights into a model's patterns of failures and biases. We show that an ensemble of such lightweight models can be used to generate insights on the performance of the black-box model.
  arXiv  Detail & Related papers  (2023-05-04T23:54:37Z)
• Learning and generalization of compositional representations of visual scenes [2.960473840509733]
  We use distributed representations of object attributes and vector operations in a vector symbolic architecture to create a full compositional description of a scene. To control the scene composition, we use artificial images composed of multiple, translated and colored MNIST digits. The output of the deep network can then be interpreted by a VSA resonator network, to extract object identity or other properties of indiviual objects.
  arXiv  Detail & Related papers  (2023-03-23T22:03:42Z)
• High Fidelity Visualization of What Your Self-Supervised Representation Knows About [22.982471878833362]
  In this work, we showcase the use of a conditional diffusion based generative model (RCDM) to visualize representations learned with self-supervised models. We demonstrate how this model's generation quality is on par with state-of-the-art generative models while being faithful to the representation used as conditioning.
  arXiv  Detail & Related papers  (2021-12-16T19:23:33Z)
• Understanding the Role of Individual Units in a Deep Neural Network [85.23117441162772]
  We present an analytic framework to systematically identify hidden units within image classification and image generation networks. First, we analyze a convolutional neural network (CNN) trained on scene classification and discover units that match a diverse set of object concepts. Second, we use a similar analytic method to analyze a generative adversarial network (GAN) model trained to generate scenes.
  arXiv  Detail & Related papers  (2020-09-10T17:59:10Z)

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.