The Representation and Recall of Interwoven Structured Knowledge in LLMs: A Geometric and Layered Analysis

• URL: http://arxiv.org/abs/2502.10871v1
• Date: Sat, 15 Feb 2025 18:08:51 GMT
• Title: The Representation and Recall of Interwoven Structured Knowledge in LLMs: A Geometric and Layered Analysis
• Authors: Ge Lei, Samuel J. Cooper,
• Abstract summary: Large language models (LLMs) represent and recall multi-associated attributes across transformer layers. intermediate layers encode factual knowledge by superimposing related attributes in overlapping spaces. later layers refine linguistic patterns and progressively separate attribute representations.
• Score: 0.0
• License:
• Abstract: This study investigates how large language models (LLMs) represent and recall multi-associated attributes across transformer layers. We show that intermediate layers encode factual knowledge by superimposing related attributes in overlapping spaces, along with effective recall even when attributes are not explicitly prompted. In contrast, later layers refine linguistic patterns and progressively separate attribute representations, optimizing task-specific outputs while appropriately narrowing attribute recall. We identify diverse encoding patterns including, for the first time, the observation of 3D spiral structures when exploring information related to the periodic table of elements. Our findings reveal a dynamic transition in attribute representations across layers, contributing to mechanistic interpretability and providing insights for understanding how LLMs handle complex, interrelated knowledge.

Related papers

• The Complexity of Learning Sparse Superposed Features with Feedback [0.9838799448847586]
  We investigate whether the underlying learned features of a model can be efficiently retrieved through feedback from an agent. We analyze the feedback complexity associated with learning a feature matrix in sparse settings. Our results establish tight bounds when the agent is permitted to construct activations and demonstrate strong upper bounds in sparse scenarios.
  arXiv  Detail & Related papers  (2025-02-08T01:54:23Z)
• Instruction-Guided Fusion of Multi-Layer Visual Features in Large Vision-Language Models [50.98559225639266]
  We investigate the contributions of visual features from different encoder layers using 18 benchmarks spanning 6 task categories. Our findings reveal that multilayer features provide complementary strengths with varying task dependencies, and uniform fusion leads to suboptimal performance. We propose the instruction-guided vision aggregator, a module that dynamically integrates multi-layer visual features based on textual instructions.
  arXiv  Detail & Related papers  (2024-12-26T05:41:31Z)
• Hybrid Discriminative Attribute-Object Embedding Network for Compositional Zero-Shot Learning [83.10178754323955]
  Hybrid Discriminative Attribute-Object Embedding (HDA-OE) network is proposed to solve the problem of complex interactions between attributes and object visual representations. To increase the variability of training data, HDA-OE introduces an attribute-driven data synthesis (ADDS) module. To further improve the discriminative ability of the model, HDA-OE introduces the subclass-driven discriminative embedding (SDDE) module. The proposed model has been evaluated on three benchmark datasets, and the results verify its effectiveness and reliability.
  arXiv  Detail & Related papers  (2024-11-28T09:50:25Z)
• Probing Ranking LLMs: Mechanistic Interpretability in Information Retrieval [22.875174888476295]
  We study the workings of state-of-the-art, fine-tuning-based passage-reranking transformer networks. Our approach involves a probing-based, layer-by-layer analysis of neurons within ranking LLMs. We identify individual or groups of known human-engineered and semantic features within the network's activations.
  arXiv  Detail & Related papers  (2024-10-24T08:20:10Z)
• Rethinking Visual Prompting for Multimodal Large Language Models with External Knowledge [76.45868419402265]
  multimodal large language models (MLLMs) have made significant strides by training on vast high-quality image-text datasets. However, the inherent difficulty in explicitly conveying fine-grained or spatially dense information in text, such as masks, poses a challenge for MLLMs. This paper proposes a new visual prompt approach to integrate fine-grained external knowledge, gleaned from specialized vision models, into MLLMs.
  arXiv  Detail & Related papers  (2024-07-05T17:43:30Z)
• Grouped Discrete Representation Guides Object-Centric Learning [18.44580501357929]
  Transformer-based Object-Centric Discrete Learning can abstract dense images or textures into sparse object-level features. We propose textitGrouped Representation (GDR) to address these issues by grouping features into attributes and indexing them with numbers.
  arXiv  Detail & Related papers  (2024-07-01T19:00:40Z)
• Dual Relation Mining Network for Zero-Shot Learning [48.89161627050706]
  We propose a Dual Relation Mining Network (DRMN) to enable effective visual-semantic interactions and learn semantic relationship among attributes for knowledge transfer. Specifically, we introduce a Dual Attention Block (DAB) for visual-semantic relationship mining, which enriches visual information by multi-level feature fusion. For semantic relationship modeling, we utilize a Semantic Interaction Transformer (SIT) to enhance the generalization of attribute representations among images.
  arXiv  Detail & Related papers  (2024-05-06T16:31:19Z)
• RefSAM: Efficiently Adapting Segmenting Anything Model for Referring Video Object Segmentation [53.4319652364256]
  This paper presents the RefSAM model, which explores the potential of SAM for referring video object segmentation. Our proposed approach adapts the original SAM model to enhance cross-modality learning by employing a lightweight Cross-RValModal. We employ a parameter-efficient tuning strategy to align and fuse the language and vision features effectively.
  arXiv  Detail & Related papers  (2023-07-03T13:21:58Z)
• Probing Linguistic Features of Sentence-Level Representations in Neural Relation Extraction [80.38130122127882]
  We introduce 14 probing tasks targeting linguistic properties relevant to neural relation extraction (RE) We use them to study representations learned by more than 40 different encoder architecture and linguistic feature combinations trained on two datasets. We find that the bias induced by the architecture and the inclusion of linguistic features are clearly expressed in the probing task performance.
  arXiv  Detail & Related papers  (2020-04-17T09:17:40Z)

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.