Can Machines Really See Objects in Images? A Study Based on Syntactic Distance and Visual Self-Referential Instances

• URL: http://arxiv.org/abs/2606.29416v1
• Date: Sun, 28 Jun 2026 14:27:40 GMT
• Title: Can Machines Really See Objects in Images? A Study Based on Syntactic Distance and Visual Self-Referential Instances
• Abstract summary: We ask whether a model can still classify correctly when such local cues provide no stable basis for distinction.<n>Experiments on ResNets and Vision Transformers reveal a consistent phase-transition phenomenon.<n>Results suggest that general intelligence may require creating new language, not reusing an existing one.
• Score: 53.65561607407989
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Can a vision model truly see an object, or does it only fit surface-level visual cues? Following Wittgenstein's view that the limits of language are the limits of the world, we view a model's recognition ability as bounded by the descriptive system it has learned. In current vision models, this system is often realized through learned feature representations that exploit local statistical cues. We therefore ask whether a model can still classify correctly when such local cues provide no stable basis for distinction. We formalize this question with syntactic distance, which measures class separability through the symmetry of the operations mapping one class to the other: positive distance exposes exploitable local features, whereas zero distance requires global semantics rather than local rules. We construct a visual self-referential task in maximum-variance binary noise: positive samples contain a closed square, while negative samples contain an otherwise identical square with one flipped boundary pixel. The two classes differ in global semantics but have zero syntactic distance, making local statistical shortcuts unreliable. Experiments on ResNets and Vision Transformers reveal a consistent phase-transition phenomenon, with accuracy collapsing to random guessing once the image scale crosses a critical point and does not recover within the tested range. Larger training sets and models only delay this collapse, while globally attentive ViTs reach it earlier. These results reveal a structural capability boundary of current architectures on global-concept tasks, suggesting that general intelligence may require creating new language, not reusing an existing one.

Related papers

• Z-FLoc: Zero-Shot Floorplan Localization via Geometric Primitives [72.63181031215858]
  We propose a zero-shot floorplan localization method that generalizes to novel environments without any retraining.<n>Our key insight is that dominant geometric primitives are ubiquitous in human-made environments.<n> Experiments on both simulated and real-world datasets show that our approach outperforms state-of-the-art learning-based methods on unseen environments.
  arXiv  Detail & Related papers  (2026-06-03T12:14:24Z)
• Seeing Isn't Knowing: Do VLMs Know When Not to Answer Spatial Questions (and Why)? [72.2500547961037]
  spatial reasoning is a fundamental capability for vision-language models (VLMs) deployed in real-world environments.<n>We introduce two types of observation challenges: (1) occlusion, which hides target information, and (2) perspective ambiguity, which produces misleading visual cues.<n>For each configuration, we design spatial questions that are answerable under clean observations but require abstention under the introduced challenges.
  arXiv  Detail & Related papers  (2026-05-28T20:44:47Z)
• Back into Plato's Cave: Examining Cross-modal Representational Convergence at Scale [40.31924374728614]
  The Platonic Representation Hypothesis suggests that neural networks trained on different modalities align and eventually converge toward the same representation of reality.<n>We show that the experimental evidence for this hypothesis is fragile and depends critically on the evaluation regime.
  arXiv  Detail & Related papers  (2026-04-20T17:56:02Z)
• TopoPerception: A Shortcut-Free Evaluation of Global Visual Perception in Large Vision-Language Models [11.7808701773328]
  Large Vision-Language Models (LVLMs) typically align visual features from an encoder with a pre-trained Large Language Model (LLM)<n>Here, we introduce TopoPerception, a benchmark that leverages topological properties to rigorously evaluate the global visual perception capabilities of LVLMs.<n>We evaluate state-of-the-art models on TopoPerception and find that even at the coarsest perceptual granularity, all models perform no better than random chance.
  arXiv  Detail & Related papers  (2025-11-14T19:45:56Z)
• Epsilon: Exploring Comprehensive Visual-Semantic Projection for Multi-Label Zero-Shot Learning [23.96220607033524]
  This paper investigates a challenging problem of zero-shot learning in the multi-label scenario (MLZSL) It is trained to recognize multiple unseen classes within a sample based on seen classes and auxiliary knowledge. We propose a novel and comprehensive visual-semantic framework for MLZSL, dubbed Epsilon, to fully make use of such properties.
  arXiv  Detail & Related papers  (2024-08-22T09:45:24Z)
• Probabilistic World Modeling with Asymmetric Distance Measure [0.3021678014343889]
  We show that learning a distance function is essential to allow planning and reasoning in the representation space. We learn an asymmetric similarity function that reflects the state reachability and allows multi-way probabilistic inference. By conditioning on a common reference state, the learned representation space allows us to discover the geometrically salient states that only a handful of paths can lead through.
  arXiv  Detail & Related papers  (2024-03-16T09:58:49Z)
• Stochastic Coherence Over Attention Trajectory For Continuous Learning In Video Streams [64.82800502603138]
  This paper proposes a novel neural-network-based approach to progressively and autonomously develop pixel-wise representations in a video stream. The proposed method is based on a human-like attention mechanism that allows the agent to learn by observing what is moving in the attended locations. Our experiments leverage 3D virtual environments and they show that the proposed agents can learn to distinguish objects just by observing the video stream.
  arXiv  Detail & Related papers  (2022-04-26T09:52:31Z)
• Synthesizing the Unseen for Zero-shot Object Detection [72.38031440014463]
  We propose to synthesize visual features for unseen classes, so that the model learns both seen and unseen objects in the visual domain. We use a novel generative model that uses class-semantics to not only generate the features but also to discriminatively separate them.
  arXiv  Detail & Related papers  (2020-10-19T12:36:11Z)
• Global-Local Bidirectional Reasoning for Unsupervised Representation Learning of 3D Point Clouds [109.0016923028653]
  We learn point cloud representation by bidirectional reasoning between the local structures and the global shape without human supervision. We show that our unsupervised model surpasses the state-of-the-art supervised methods on both synthetic and real-world 3D object classification datasets.
  arXiv  Detail & Related papers  (2020-03-29T08:26:08Z)

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.