Talk in Pieces, See in Whole: Disentangling and Hierarchical Aggregating Representations for Language-based Object Detection

• URL: http://arxiv.org/abs/2509.24192v1
• Date: Mon, 29 Sep 2025 02:14:26 GMT
• Title: Talk in Pieces, See in Whole: Disentangling and Hierarchical Aggregating Representations for Language-based Object Detection
• Authors: Sojung An, Kwanyong Park, Yong Jae Lee, Donghyun Kim,
• Abstract summary: We propose restructuring linguistic representations according to the hierarchical relations within sentences for language-based object detection.<n>A key insight is the necessity of disentangling textual tokens into core components-objects, attributes, and relations ("talk in pieces")-and subsequently aggregating them into hierarchically structured sentence-level representations.<n> Experimental results under the OmniLabel benchmark show a 24% performance improvement, demonstrating the importance of linguistic compositionality.
• Score: 39.748035737067745
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: While vision-language models (VLMs) have made significant progress in multimodal perception (e.g., open-vocabulary object detection) with simple language queries, state-of-the-art VLMs still show limited ability to perceive complex queries involving descriptive attributes and relational clauses. Our in-depth analysis shows that these limitations mainly stem from text encoders in VLMs. Such text encoders behave like bags-of-words and fail to separate target objects from their descriptive attributes and relations in complex queries, resulting in frequent false positives. To address this, we propose restructuring linguistic representations according to the hierarchical relations within sentences for language-based object detection. A key insight is the necessity of disentangling textual tokens into core components-objects, attributes, and relations ("talk in pieces")-and subsequently aggregating them into hierarchically structured sentence-level representations ("see in whole"). Building on this principle, we introduce the TaSe framework with three main contributions: (1) a hierarchical synthetic captioning dataset spanning three tiers from category names to descriptive sentences; (2) Talk in Pieces, the three-component disentanglement module guided by a novel disentanglement loss function, transforms text embeddings into subspace compositions; and (3) See in Whole, which learns to aggregate disentangled components into hierarchically structured embeddings with the guide of proposed hierarchical objectives. The proposed TaSe framework strengthens the inductive bias of hierarchical linguistic structures, resulting in fine-grained multimodal representations for language-based object detection. Experimental results under the OmniLabel benchmark show a 24% performance improvement, demonstrating the importance of linguistic compositionality.

Related papers

• MoDora: Tree-Based Semi-Structured Document Analysis System [62.01015188258797]
  Semi-structured documents integrate diverse interleaved data elements arranged in various and often irregular layouts.<n>MoDora is an LLM-powered system for semi-structured document analysis.<n> Experiments show MoDora outperforms baselines by 5.97%-61.07% in accuracy.
  arXiv  Detail & Related papers  (2026-02-26T14:48:49Z)
• Understanding Subword Compositionality of Large Language Models [42.51978887170929]
  Large language models (LLMs) take sequences of subwords as input, requiring them to compose subword representations.<n>We present a comprehensive set of experiments to probe how LLMs compose subword information.
  arXiv  Detail & Related papers  (2025-08-25T12:16:56Z)
• Composed Object Retrieval: Object-level Retrieval via Composed Expressions [71.47650333199628]
  Composed Object Retrieval (COR) is a brand-new task that goes beyond image-level retrieval to achieve object-level precision.<n>We construct COR127K, the first large-scale COR benchmark that contains 127,166 retrieval triplets with various semantic transformations in 408 categories.<n>We also present CORE, a unified end-to-end model that integrates reference region encoding, adaptive visual-textual interaction, and region-level contrastive learning.
  arXiv  Detail & Related papers  (2025-08-06T13:11:40Z)
• DISRetrieval: Harnessing Discourse Structure for Long Document Retrieval [51.89673002051528]
  DISRetrieval is a novel hierarchical retrieval framework that leverages linguistic discourse structure to enhance long document understanding.<n>Our studies confirm that discourse structure significantly enhances retrieval effectiveness across different document lengths and query types.
  arXiv  Detail & Related papers  (2025-05-26T14:45:12Z)
• SRFUND: A Multi-Granularity Hierarchical Structure Reconstruction Benchmark in Form Understanding [55.48936731641802]
  We present the SRFUND, a hierarchically structured multi-task form understanding benchmark. SRFUND provides refined annotations on top of the original FUNSD and XFUND datasets. The dataset includes eight languages including English, Chinese, Japanese, German, French, Spanish, Italian, and Portuguese.
  arXiv  Detail & Related papers  (2024-06-13T02:35:55Z)
• 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)
• Can Linguistic Knowledge Improve Multimodal Alignment in Vision-Language Pretraining? [34.609984453754656]
  We aim to elucidate the impact of comprehensive linguistic knowledge, including semantic expression and syntactic structure, on multimodal alignment. Specifically, we design and release the SNARE, the first large-scale multimodal alignment probing benchmark.
  arXiv  Detail & Related papers  (2023-08-24T16:17:40Z)
• Multi-Modal Classifiers for Open-Vocabulary Object Detection [104.77331131447541]
  The goal of this paper is open-vocabulary object detection (OVOD) We adopt a standard two-stage object detector architecture. We explore three ways via: language descriptions, image exemplars, or a combination of the two.
  arXiv  Detail & Related papers  (2023-06-08T18:31:56Z)
• Few-Shot Nested Named Entity Recognition [4.8693196802491405]
  This paper is the first one dedicated to studying the few-shot nested NER task. We propose a Biaffine-based Contrastive Learning (BCL) framework to learn contextual dependency to distinguish nested entities. The BCL outperformed three baseline models on the 1-shot and 5-shot tasks in terms of F1 score.
  arXiv  Detail & Related papers  (2022-12-02T03:42:23Z)
• Hierarchical Modular Network for Video Captioning [162.70349114104107]
  We propose a hierarchical modular network to bridge video representations and linguistic semantics from three levels before generating captions. The proposed method performs favorably against the state-of-the-art models on the two widely-used benchmarks: MSVD 104.0% and MSR-VTT 51.5% in CIDEr score.
  arXiv  Detail & Related papers  (2021-11-24T13:07:05Z)
• A Self-supervised Representation Learning of Sentence Structure for Authorship Attribution [3.5991811164452923]
  We propose a self-supervised framework for learning structural representations of sentences. We evaluate the learned structural representations of sentences using different probing tasks, and subsequently utilize them in the authorship attribution task.
  arXiv  Detail & Related papers  (2020-10-14T02:57:10Z)

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.