Breaking Token Into Concepts: Exploring Extreme Compression in Token Representation Via Compositional Shared Semantics

• URL: http://arxiv.org/abs/2509.17737v2
• Date: Tue, 23 Sep 2025 08:18:07 GMT
• Title: Breaking Token Into Concepts: Exploring Extreme Compression in Token Representation Via Compositional Shared Semantics
• Abstract summary: We investigate whether tokens can be more effectively represented through a compositional structure that accumulates diverse semantic facets.<n>We propose Aggregate Semantic Grouping (ASG), a novel approach leveraging Product Quantization (PQ)<n>Our findings demonstrate that representing tokens compositionally via ASG achieves extreme compression in embedding parameters.<n>These results validate the principle that tokens can be effectively modeled as combinations of shared semantic building blocks.
• Score: 4.822201556735661
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Standard language models employ unique, monolithic embeddings for each token, potentially limiting their ability to capture the multifaceted nature of word meanings. We investigate whether tokens can be more effectively represented through a compositional structure that accumulates diverse semantic facets. To explore this, we propose Aggregate Semantic Grouping (ASG), a novel approach leveraging Product Quantization (PQ). We apply ASG to standard transformer architectures (mBERT, XLM-R, mT5) and evaluate this representational scheme across diverse tasks (NLI, NER, QA), as well as a biomedical domain-specific benchmark (BC5CDR) using BioBERT. Our findings demonstrate that representing tokens compositionally via ASG achieves extreme compression in embedding parameters (0.4--0.5\%) while maintaining $>$95\% task performance relative to the base model, even in generative tasks and extends to both cross lingual transfer and domain-specific settings. These results validate the principle that tokens can be effectively modeled as combinations of shared semantic building blocks. ASG offers a simple yet concrete method for achieving this, showcasing how compositional representations can capture linguistic richness while enabling compact yet semantically rich models.

Related papers

• Tokenization, Fusion and Decoupling: Bridging the Granularity Mismatch Between Large Language Models and Knowledge Graphs [20.946228883628013]
  We propose KGT, a novel framework that uses dedicated entity tokens to enable efficient, full-space prediction.<n>We first introduce specialized tokenization to construct feature representations at the level of dedicated entity tokens.<n>We then fuse pre-trained structural and textual features into these unified embeddings via a relation-guided gating mechanism.
  arXiv  Detail & Related papers  (2026-02-26T07:20:40Z)
• Unleash the Potential of Long Semantic IDs for Generative Recommendation [5.6264583086973685]
  ACERec is a novel framework that decouples the gap between fine-grained tokenization and efficient sequential modeling.<n>It consistently outperforms state-of-the-art baselines on six real-world benchmarks.
  arXiv  Detail & Related papers  (2026-02-14T03:15:31Z)
• CSMCIR: CoT-Enhanced Symmetric Alignment with Memory Bank for Composed Image Retrieval [54.15776146365823]
  Composed Image Retrieval (CIR) enables users to search for target images using both a reference image and manipulation text.<n>We propose CSMCIR, a unified representation framework that achieves efficient query-target alignment through three synergistic components.
  arXiv  Detail & Related papers  (2026-01-07T09:21:38Z)
• MergeDNA: Context-aware Genome Modeling with Dynamic Tokenization through Token Merging [65.07273789940116]
  This paper introduces a hierarchical architecture that jointly optimize a dynamic genomic tokenizer and latent Transformers with context-aware pre-training tasks.<n> MergeDNA achieves superior performance on three popular DNA benchmarks and several multi-omics tasks with fine-tuning or zero-shot evaluation.
  arXiv  Detail & Related papers  (2025-11-17T19:27:41Z)
• Text4Seg++: Advancing Image Segmentation via Generative Language Modeling [52.07442359419673]
  We propose a novel text-as-mask paradigm that casts image segmentation as a text generation problem.<n>Key innovation is semantic descriptors, a new textual representation of segmentation masks.<n>Experiments on natural and remote sensing datasets show that Text4Seg++ consistently outperforms state-of-the-art models.
  arXiv  Detail & Related papers  (2025-09-08T04:07:14Z)
• Inceptive Transformers: Enhancing Contextual Representations through Multi-Scale Feature Learning Across Domains and Languages [3.294155819837931]
  transformer models compress information from all tokens in a sequence into a single [/] token to represent global context.<n>This approach risks diluting fine-grained or hierarchical features, leading to information loss in downstream tasks where local patterns are important.<n>We propose an inception-style 1-D convolution module that sits on top of the transformer layer and augments token representations with multi-scale local features.
  arXiv  Detail & Related papers  (2025-05-26T19:59:22Z)
• Learning Multi-Aspect Item Palette: A Semantic Tokenization Framework for Generative Recommendation [55.99632509895994]
  We introduce LAMIA, a novel approach for multi-aspect semantic tokenization.<n>Unlike RQ-VAE, which uses a single embedding, LAMIA learns an item palette''--a collection of independent and semantically parallel embeddings.<n>Our results demonstrate significant improvements in recommendation accuracy over existing methods.
  arXiv  Detail & Related papers  (2024-09-11T13:49:48Z)
• Subobject-level Image Tokenization [60.80949852899857]
  Patch-based image tokenization ignores the morphology of the visual world.<n>Inspired by subword tokenization, we introduce subobject-level adaptive token segmentation.<n>We show that subobject tokenization enables faster convergence and better generalization while using fewer visual tokens.
  arXiv  Detail & Related papers  (2024-02-22T06:47:44Z)
• Uncovering Prototypical Knowledge for Weakly Open-Vocabulary Semantic Segmentation [59.37587762543934]
  This paper studies the problem of weakly open-vocabulary semantic segmentation (WOVSS) Existing methods suffer from a granularity inconsistency regarding the usage of group tokens. We propose the prototypical guidance network (PGSeg) that incorporates multi-modal regularization.
  arXiv  Detail & Related papers  (2023-10-29T13:18:00Z)
• Improving Generalization in Language Model-Based Text-to-SQL Semantic Parsing: Two Simple Semantic Boundary-Based Techniques [14.634536051274468]
  We introduce a token preprocessing method to preserve the semantic boundaries of tokens produced by LM tokenizers. At the sequence level, we propose to use special tokens to mark the boundaries of components aligned between input and output. Our experimental results on two text-to- semantic parsing datasets show that our token preprocessing, although simple, can substantially improve the LM performance on both types of generalization.
  arXiv  Detail & Related papers  (2023-05-27T06:09:03Z)
• X2Parser: Cross-Lingual and Cross-Domain Framework for Task-Oriented Compositional Semantic Parsing [51.81533991497547]
  Task-oriented compositional semantic parsing (TCSP) handles complex nested user queries. We present X2 compared a transferable Cross-lingual and Cross-domain for TCSP. We propose to predict flattened intents and slots representations separately and cast both prediction tasks into sequence labeling problems.
  arXiv  Detail & Related papers  (2021-06-07T16:40:05Z)
• Structured Prediction as Translation between Augmented Natural Languages [109.50236248762877]
  We propose a new framework, Translation between Augmented Natural Languages (TANL), to solve many structured prediction language tasks. Instead of tackling the problem by training task-specific discriminatives, we frame it as a translation task between augmented natural languages. Our approach can match or outperform task-specific models on all tasks, and in particular, achieves new state-of-the-art results on joint entity and relation extraction.
  arXiv  Detail & Related papers  (2021-01-14T18:32:21Z)

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.