Not All Splits Are Equal: Rethinking Attribute Generalization Across Unrelated Categories

• URL: http://arxiv.org/abs/2509.06998v1
• Date: Thu, 04 Sep 2025 17:52:22 GMT
• Title: Not All Splits Are Equal: Rethinking Attribute Generalization Across Unrelated Categories
• Authors: Liviu Nicolae Fircă, Antonio Bărbălau, Dan Oneata, Elena Burceanu,
• Abstract summary: This work presents the first explicit evaluation for the robustness of the attribute prediction task under such conditions.<n>We introduce train-test split strategies that progressively reduce correlation between training and test sets.<n>Results show a sharp drop in performance as the correlation between training and test categories decreases.
• Score: 7.048738972394742
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Can models generalize attribute knowledge across semantically and perceptually dissimilar categories? While prior work has addressed attribute prediction within narrow taxonomic or visually similar domains, it remains unclear whether current models can abstract attributes and apply them to conceptually distant categories. This work presents the first explicit evaluation for the robustness of the attribute prediction task under such conditions, testing whether models can correctly infer shared attributes between unrelated object types: e.g., identifying that the attribute "has four legs" is common to both "dogs" and "chairs". To enable this evaluation, we introduce train-test split strategies that progressively reduce correlation between training and test sets, based on: LLM-driven semantic grouping, embedding similarity thresholding, embedding-based clustering, and supercategory-based partitioning using ground-truth labels. Results show a sharp drop in performance as the correlation between training and test categories decreases, indicating strong sensitivity to split design. Among the evaluated methods, clustering yields the most effective trade-off, reducing hidden correlations while preserving learnability. These findings offer new insights into the limitations of current representations and inform future benchmark construction for attribute reasoning.

Related papers

• Bridging the Semantic Gap for Categorical Data Clustering via Large Language Models [64.58262227709842]
  ARISE (Attention-weighted Representation with Integrated Semantic Embeddings) is presented.<n>It builds semantic-aware representations that complement the metric space of categorical data for accurate clustering.<n>Experiments on eight benchmark datasets demonstrate consistent improvements over seven representative counterparts.
  arXiv  Detail & Related papers  (2026-01-03T11:37:46Z)
• Break the Tie: Learning Cluster-Customized Category Relationships for Categorical Data Clustering [51.11677202873771]
  Categorical attributes with qualitative values are ubiquitous in cluster analysis of real datasets.<n>Unlike the Euclidean distance of numerical attributes, the categorical attributes lack well-defined relationships of their possible values.<n>This paper breaks the intrinsic relationship tie of attribute categories and learns customized distance metrics suitable for flexibly revealing various cluster distributions.
  arXiv  Detail & Related papers  (2025-11-12T06:57:24Z)
• Learning Robust Classifiers with Self-Guided Spurious Correlation Mitigation [26.544938760265136]
  Deep neural classifiers rely on spurious correlations between spurious attributes of inputs and targets to make predictions. We propose a self-guided spurious correlation mitigation framework. We show that training the classifier to distinguish different prediction behaviors reduces its reliance on spurious correlations without knowing them a priori.
  arXiv  Detail & Related papers  (2024-05-06T17:12:21Z)
• AttrSeg: Open-Vocabulary Semantic Segmentation via Attribute Decomposition-Aggregation [33.25304533086283]
  Open-vocabulary semantic segmentation is a challenging task that requires segmenting novel object categories at inference time. Recent studies have explored vision-language pre-training to handle this task, but suffer from unrealistic assumptions in practical scenarios. This work proposes a novel attribute decomposition-aggregation framework, AttrSeg, inspired by human cognition in understanding new concepts.
  arXiv  Detail & Related papers  (2023-08-31T19:34:09Z)
• Learning Debiased and Disentangled Representations for Semantic Segmentation [52.35766945827972]
  We propose a model-agnostic and training scheme for semantic segmentation. By randomly eliminating certain class information in each training iteration, we effectively reduce feature dependencies among classes. Models trained with our approach demonstrate strong results on multiple semantic segmentation benchmarks.
  arXiv  Detail & Related papers  (2021-10-31T16:15:09Z)
• Multi-Objective Few-shot Learning for Fair Classification [23.05869193599414]
  We propose a framework for mitigating the disparities of the predicted classes with respect to secondary attributes within the data. Our proposed method involves learning a multi-objective function that in addition to learning the primary objective of predicting the primary class labels from the data, also employs a clustering-based to minimize the disparities of the class label distribution with respect to the cluster memberships.
  arXiv  Detail & Related papers  (2021-10-05T11:28:58Z)
• Contrastive Learning for Fair Representations [50.95604482330149]
  Trained classification models can unintentionally lead to biased representations and predictions. Existing debiasing methods for classification models, such as adversarial training, are often expensive to train and difficult to optimise. We propose a method for mitigating bias by incorporating contrastive learning, in which instances sharing the same class label are encouraged to have similar representations.
  arXiv  Detail & Related papers  (2021-09-22T10:47:51Z)
• Intersection Regularization for Extracting Semantic Attributes [72.53481390411173]
  We consider the problem of supervised classification, such that the features that the network extracts match an unseen set of semantic attributes. For example, when learning to classify images of birds into species, we would like to observe the emergence of features that zoologists use to classify birds. We propose training a neural network with discrete top-level activations, which is followed by a multi-layered perceptron (MLP) and a parallel decision tree.
  arXiv  Detail & Related papers  (2021-03-22T14:32:44Z)
• Dynamic Semantic Matching and Aggregation Network for Few-shot Intent Detection [69.2370349274216]
  Few-shot Intent Detection is challenging due to the scarcity of available annotated utterances. Semantic components are distilled from utterances via multi-head self-attention. Our method provides a comprehensive matching measure to enhance representations of both labeled and unlabeled instances.
  arXiv  Detail & Related papers  (2020-10-06T05:16:38Z)
• Identifying Spurious Correlations for Robust Text Classification [9.457737910527829]
  We propose a method to distinguish spurious and genuine correlations in text classification. We use features derived from treatment effect estimators to distinguish spurious correlations from "genuine" ones. Experiments on four datasets suggest that using this approach to inform feature selection also leads to more robust classification.
  arXiv  Detail & Related papers  (2020-10-06T03:49:22Z)
• Commonality-Parsing Network across Shape and Appearance for Partially Supervised Instance Segmentation [71.59275788106622]
  We propose to learn the underlying class-agnostic commonalities that can be generalized from mask-annotated categories to novel categories. Our model significantly outperforms the state-of-the-art methods on both partially supervised setting and few-shot setting for instance segmentation on COCO dataset.
  arXiv  Detail & Related papers  (2020-07-24T07:23:44Z)
• On Irrelevance of Attributes in Flexible Prediction [0.0]
  This paper analyses properties of conceptual hierarchy obtained via incremental concept formation method called "flexible prediction" The impact of selection of simple and combined attributes, of scaling and of distribution of individual attributes and of correlation strengths among them is investigated.
  arXiv  Detail & Related papers  (2020-05-25T08:41:48Z)
• Don't Judge an Object by Its Context: Learning to Overcome Contextual Bias [113.44471186752018]
  Existing models often leverage co-occurrences between objects and their context to improve recognition accuracy. This work focuses on addressing such contextual biases to improve the robustness of the learnt feature representations.
  arXiv  Detail & Related papers  (2020-01-09T18:31:55Z)

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.