Back into Plato's Cave: Examining Cross-modal Representational Convergence at Scale

• URL: http://arxiv.org/abs/2604.18572v1
• Date: Mon, 20 Apr 2026 17:56:02 GMT
• Title: Back into Plato's Cave: Examining Cross-modal Representational Convergence at Scale
• Abstract summary: 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.
• Score: 40.31924374728614
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The Platonic Representation Hypothesis suggests that neural networks trained on different modalities (e.g., text and images) align and eventually converge toward the same representation of reality. If true, this has significant implications for whether modality choice matters at all. We show that the experimental evidence for this hypothesis is fragile and depends critically on the evaluation regime. Alignment is measured using mutual nearest neighbors on small datasets ($\approx$1K samples) and degrades substantially as the dataset is scaled to millions of samples. The alignment that remains between model representations reflects coarse semantic overlap rather than consistent fine-grained structure. Moreover, the evaluations in Huh et al. are done in a one-to-one image-caption setting, a constraint that breaks down in realistic many-to-many settings and further reduces alignment. We also find that the reported trend of stronger language models increasingly aligning with vision does not appear to hold for newer models. Overall, our findings suggest that the current evidence for cross-modal representational convergence is considerably weaker than subsequent works have taken it to be. Models trained on different modalities may learn equally rich representations of the world, just not the same one.

Related papers

• Mapping and Measuring the Behavioral Evolution of Large Language Models [28.603443859513153]
  Benchmark leaderboards summarize how well a language model performs, but not how its behavior relates to that of other models or changes across generations.<n>We characterize the output behavior of 32 models from six families using their responses to a shared bank of 10,000 prompts.
  arXiv  Detail & Related papers  (2026-08-11T15:07:28Z)
• Can Machines Really See Objects in Images? A Study Based on Syntactic Distance and Visual Self-Referential Instances [53.65561607407989]
  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.
  arXiv  Detail & Related papers  (2026-06-28T14:27:40Z)
• Unveiling the Visual Counting Bottleneck in Vision-Language Models [49.591496870141846]
  This work deconstructs visual counting into three cognitive stages: visual individuation, magnitude awareness, and symbolic mapping.<n>Using synthetic Go boards and linear probes, we demonstrate that visual backbones maintain robust, linearly separable representations of quantity well into the extrapolation regime.<n>We pinpoint the collapse to the symbolic mapping stage, where the model fails to project valid visual magnitudes onto symbolic tokens.
  arXiv  Detail & Related papers  (2026-05-28T16:20:29Z)
• Revisiting the Platonic Representation Hypothesis: An Aristotelian View [3.647057737530591]
  We show that the existing metrics used to measure representational similarity are confounded by network scale.<n>We introduce a permutation-based null-calibration framework that transforms any representational similarity metric into a calibrated score with statistical guarantees.<n>We propose the Aristotelian Representation Hypothesis: representations in neural networks are converging to shared local neighborhood relationships.
  arXiv  Detail & Related papers  (2026-02-16T06:01:23Z)
• When Does Closeness in Distribution Imply Representational Similarity? An Identifiability Perspective [9.578534178372829]
  We prove that a small Kullback--Leibler divergence between the model distributions does not guarantee that the corresponding representations are similar.<n>We then define a distributional distance for which closeness implies representational similarity.<n>In synthetic experiments, we find that wider networks learn distributions which are closer with respect to our distance and have more similar representations.
  arXiv  Detail & Related papers  (2025-06-04T09:44:22Z)
• Bridging Critical Gaps in Convergent Learning: How Representational Alignment Evolves Across Layers, Training, and Distribution Shifts [1.9458156037869137]
  convergent learning is the degree to which neural systems arrive at similar internal representations.<n>We present a large-scale audit of convergent learning spanning dozens of vision models and thousands of layer-pair comparisons.<n>Findings fill critical gaps in our understanding of representational convergence, with implications for neuroscience and AI.
  arXiv  Detail & Related papers  (2025-02-26T00:04:24Z)
• Improving Network Interpretability via Explanation Consistency Evaluation [56.14036428778861]
  We propose a framework that acquires more explainable activation heatmaps and simultaneously increase the model performance. Specifically, our framework introduces a new metric, i.e., explanation consistency, to reweight the training samples adaptively in model learning. Our framework then promotes the model learning by paying closer attention to those training samples with a high difference in explanations.
  arXiv  Detail & Related papers  (2024-08-08T17:20:08Z)
• The Platonic Representation Hypothesis [35.16414255187554]
  We argue that representations in AI models, particularly deep networks, are converging. As vision models and language models get larger, they measure distance between datapoints in a more and more alike way. We hypothesize that this convergence is driving toward a shared statistical model of reality, akin to Plato's concept of an ideal reality.
  arXiv  Detail & Related papers  (2024-05-13T17:58:30Z)
• Chaos is a Ladder: A New Theoretical Understanding of Contrastive Learning via Augmentation Overlap [64.60460828425502]
  We propose a new guarantee on the downstream performance of contrastive learning. Our new theory hinges on the insight that the support of different intra-class samples will become more overlapped under aggressive data augmentations. We propose an unsupervised model selection metric ARC that aligns well with downstream accuracy.
  arXiv  Detail & Related papers  (2022-03-25T05:36:26Z)
• Perceptual Score: What Data Modalities Does Your Model Perceive? [73.75255606437808]
  We introduce the perceptual score, a metric that assesses the degree to which a model relies on the different subsets of the input features. We find that recent, more accurate multi-modal models for visual question-answering tend to perceive the visual data less than their predecessors. Using the perceptual score also helps to analyze model biases by decomposing the score into data subset contributions.
  arXiv  Detail & Related papers  (2021-10-27T12:19:56Z)
• Finding Geometric Models by Clustering in the Consensus Space [61.65661010039768]
  We propose a new algorithm for finding an unknown number of geometric models, e.g., homographies. We present a number of applications where the use of multiple geometric models improves accuracy. These include pose estimation from multiple generalized homographies; trajectory estimation of fast-moving objects.
  arXiv  Detail & Related papers  (2021-03-25T14:35:07Z)
• A Sober Look at the Unsupervised Learning of Disentangled Representations and their Evaluation [63.042651834453544]
  We show that the unsupervised learning of disentangled representations is impossible without inductive biases on both the models and the data. We observe that while the different methods successfully enforce properties "encouraged" by the corresponding losses, well-disentangled models seemingly cannot be identified without supervision. Our results suggest that future work on disentanglement learning should be explicit about the role of inductive biases and (implicit) supervision.
  arXiv  Detail & Related papers  (2020-10-27T10:17:15Z)

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.