Tackling the Abstraction and Reasoning Corpus with Vision Transformers: the Importance of 2D Representation, Positions, and Objects

• URL: http://arxiv.org/abs/2410.06405v1
• Date: Tue, 8 Oct 2024 22:25:34 GMT
• Title: Tackling the Abstraction and Reasoning Corpus with Vision Transformers: the Importance of 2D Representation, Positions, and Objects
• Authors: Wenhao Li, Yudong Xu, Scott Sanner, Elias Boutros Khalil,
• Abstract summary: We show that a Vision Transformer (ViT) fails dramatically on most ARC tasks even when trained on one million examples per task. We propose ViTARC, a ViT-style architecture that unlocks some of the visual reasoning capabilities required by the ARC. Our task-specific ViTARC models achieve a test solve rate close to 100% on more than half of the 400 public ARC tasks.
• Score: 31.926206783846144
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The Abstraction and Reasoning Corpus (ARC) is a popular benchmark focused on visual reasoning in the evaluation of Artificial Intelligence systems. In its original framing, an ARC task requires solving a program synthesis problem over small 2D images using a few input-output training pairs. In this work, we adopt the recently popular data-driven approach to the ARC and ask whether a Vision Transformer (ViT) can learn the implicit mapping, from input image to output image, that underlies the task. We show that a ViT -- otherwise a state-of-the-art model for images -- fails dramatically on most ARC tasks even when trained on one million examples per task. This points to an inherent representational deficiency of the ViT architecture that makes it incapable of uncovering the simple structured mappings underlying the ARC tasks. Building on these insights, we propose ViTARC, a ViT-style architecture that unlocks some of the visual reasoning capabilities required by the ARC. Specifically, we use a pixel-level input representation, design a spatially-aware tokenization scheme, and introduce a novel object-based positional encoding that leverages automatic segmentation, among other enhancements. Our task-specific ViTARC models achieve a test solve rate close to 100% on more than half of the 400 public ARC tasks strictly through supervised learning from input-output grids. This calls attention to the importance of imbuing the powerful (Vision) Transformer with the correct inductive biases for abstract visual reasoning that are critical even when the training data is plentiful and the mapping is noise-free. Hence, ViTARC provides a strong foundation for future research in visual reasoning using transformer-based architectures.

Related papers

• Fibottention: Inceptive Visual Representation Learning with Diverse Attention Across Heads [10.169639612525643]
  Visual perception tasks are predominantly solved by ViT, despite their effectiveness. Despite their effectiveness, ViT encounters a computational bottleneck due to the complexity of computing self-attention. We propose Fibottention architecture, which approximating self-attention that is built upon.
  arXiv  Detail & Related papers  (2024-06-27T17:59:40Z)
• Do Vision-Language Transformers Exhibit Visual Commonsense? An Empirical Study of VCR [51.72751335574947]
  Visual Commonsense Reasoning (VCR) calls for explanatory reasoning behind question answering over visual scenes. Progress on the benchmark dataset stems largely from the recent advancement of Vision-Language Transformers (VL Transformers) This paper posits that the VL Transformers do not exhibit visual commonsense, which is the key to VCR.
  arXiv  Detail & Related papers  (2024-05-27T08:26:58Z)
• TaskCLIP: Extend Large Vision-Language Model for Task Oriented Object Detection [23.73648235283315]
  Task-oriented object detection aims to find objects suitable for accomplishing specific tasks. Recent solutions are mainly all-in-one models. We propose TaskCLIP, a more natural two-stage design composed of general object detection and task-guided object selection.
  arXiv  Detail & Related papers  (2024-03-12T22:33:02Z)
• UPOCR: Towards Unified Pixel-Level OCR Interface [36.966005829678124]
  We propose UPOCR, a simple-yet-effective generalist model for Unified Pixel-level OCR interface. Specifically, the UPOCR unifies the paradigm of diverse OCR tasks as image-to-image transformation and the architecture as a vision Transformer (ViT)-based encoder-decoder. Experiments are conducted on three pixel-level OCR tasks including text removal, text segmentation, and tampered text detection.
  arXiv  Detail & Related papers  (2023-12-05T11:53:17Z)
• Vision-by-Language for Training-Free Compositional Image Retrieval [78.60509831598745]
  Compositional Image Retrieval (CIR) aims to retrieve the relevant target image in a database. Recent research sidesteps this need by using large-scale vision-language models (VLMs) We propose to tackle CIR in a training-free manner via Vision-by-Language (CIReVL)
  arXiv  Detail & Related papers  (2023-10-13T17:59:38Z)
• Solving Reasoning Tasks with a Slot Transformer [7.966351917016229]
  We present the Slot Transformer, an architecture that leverages slot attention, transformers and iterative variational inference on video scene data to infer representations. We evaluate the effectiveness of key components of the architecture, the model's representational capacity and its ability to predict from incomplete input.
  arXiv  Detail & Related papers  (2022-10-20T16:40:30Z)
• mPLUG: Effective and Efficient Vision-Language Learning by Cross-modal Skip-connections [104.14624185375897]
  mPLUG is a new vision-language foundation model for both cross-modal understanding and generation. It achieves state-of-the-art results on a wide range of vision-language downstream tasks, such as image captioning, image-text retrieval, visual grounding and visual question answering.
  arXiv  Detail & Related papers  (2022-05-24T11:52:06Z)
• Multitask AET with Orthogonal Tangent Regularity for Dark Object Detection [84.52197307286681]
  We propose a novel multitask auto encoding transformation (MAET) model to enhance object detection in a dark environment. In a self-supervision manner, the MAET learns the intrinsic visual structure by encoding and decoding the realistic illumination-degrading transformation. We have achieved the state-of-the-art performance using synthetic and real-world datasets.
  arXiv  Detail & Related papers  (2022-05-06T16:27:14Z)
• Efficient Self-supervised Vision Transformers for Representation Learning [86.57557009109411]
  We show that multi-stage architectures with sparse self-attentions can significantly reduce modeling complexity. We propose a new pre-training task of region matching which allows the model to capture fine-grained region dependencies. Our results show that combining the two techniques, EsViT achieves 81.3% top-1 on the ImageNet linear probe evaluation.
  arXiv  Detail & Related papers  (2021-06-17T19:57:33Z)
• Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision [57.031588264841]
  We leverage a noisy dataset of over one billion image alt-text pairs, obtained without expensive filtering or post-processing steps. A simple dual-encoder architecture learns to align visual and language representations of the image and text pairs using a contrastive loss. We show that the scale of our corpus can make up for its noise and leads to state-of-the-art representations even with such a simple learning scheme.
  arXiv  Detail & Related papers  (2021-02-11T10:08:12Z)

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.