Tracing Relational Knowledge Recall in Large Language Models

• URL: http://arxiv.org/abs/2604.19934v2
• Date: Thu, 23 Apr 2026 11:59:25 GMT
• Title: Tracing Relational Knowledge Recall in Large Language Models
• Abstract summary: We study how large language models recall knowledge during text generation.<n>We focus on identifying latent representations suitable for relation classification via linear probes.<n>We show that per-head attention contributions to the residual stream are comparatively strong features for linear relation classification.
• Score: 4.990228412613982
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We study how large language models recall relational knowledge during text generation, with a focus on identifying latent representations suitable for relation classification via linear probes. Prior work shows how attention heads and MLPs interact to resolve subject, predicate, and object, but it remains unclear which representations support faithful linear relation classification and why some relation types are easier to capture linearly than others. We systematically evaluate different latent representations derived from attention head and MLP contributions, showing that per-head attention contributions to the residual stream are comparatively strong features for linear relation classification. Feature attribution analyses of the trained probes, as well as characteristics of the different relation types, reveal clear correlations between probe accuracy and relation specificity, entity connectedness, and how distributed the signal on which the probe relies is across attention heads. Finally, we show how token-level feature attribution of probe predictions can be used to reveal probe behavior in further detail.

Related papers

• K-Models: a Flexible and Interpretable Method for Ordinal Clustering with Application to Antigen-Antibody Interaction Profiles [65.21921642886367]
  K-Models is a novel framework that integrates ordinal constraints and estimates key underlying elements of the random process generating the observed functional profiles.<n>The proposed method is evaluated through simulations and real-world applications.
  arXiv  Detail & Related papers  (2026-05-14T13:35:44Z)
• MINERVA: Mutual Information Neural Estimation for Supervised Feature Selection [39.57737590420284]
  We introduce a novel approach to supervised feature selection based on neural estimation of mutual information between features and targets.<n>We paramaterize the approximation of mutual information with neural networks and perform feature selection using a carefully designed loss function augmented with sparsity-inducing regularizers.<n>Our method is implemented in a two-stage process to decouple representation learning from feature selection, ensuring better generalization and a more accurate expression of feature importance.
  arXiv  Detail & Related papers  (2025-10-02T23:09:06Z)
• Hallucination Detection in LLMs with Topological Divergence on Attention Graphs [60.83579255387347]
  Hallucination, i.e., generating factually incorrect content, remains a critical challenge for large language models.<n>We introduce TOHA, a TOpology-based HAllucination detector in the RAG setting.
  arXiv  Detail & Related papers  (2025-04-14T10:06:27Z)
• Understanding Probe Behaviors through Variational Bounds of Mutual Information [53.520525292756005]
  We provide guidelines for linear probing by constructing a novel mathematical framework leveraging information theory. First, we connect probing with the variational bounds of mutual information (MI) to relax the probe design, equating linear probing with fine-tuning. We show that the intermediate representations can have the biggest MI estimate because of the tradeoff between better separability and decreasing MI.
  arXiv  Detail & Related papers  (2023-12-15T18:38:18Z)
• Learning Complete Topology-Aware Correlations Between Relations for Inductive Link Prediction [121.65152276851619]
  We show that semantic correlations between relations are inherently edge-level and entity-independent. We propose a novel subgraph-based method, namely TACO, to model Topology-Aware COrrelations between relations. To further exploit the potential of RCN, we propose Complete Common Neighbor induced subgraph.
  arXiv  Detail & Related papers  (2023-09-20T08:11:58Z)
• Document-level Relation Extraction with Relation Correlations [15.997345900917058]
  Document-level relation extraction faces two overlooked challenges: long-tail problem and multi-label problem. We analyze the co-occurrence correlation of relations, and introduce it into DocRE task for the first time.
  arXiv  Detail & Related papers  (2022-12-20T11:17:52Z)
• Relation-dependent Contrastive Learning with Cluster Sampling for Inductive Relation Prediction [30.404149577013595]
  We introduce Relation-dependent Contrastive Learning (ReCoLe) for inductive relation prediction. GNN-based encoder is optimized by contrastive learning, which ensures satisfactory performance on long-tail relations. Experimental results suggest that ReCoLe outperforms state-of-the-art methods on commonly used inductive datasets.
  arXiv  Detail & Related papers  (2022-11-22T13:30:49Z)
• On Neural Architecture Inductive Biases for Relational Tasks [76.18938462270503]
  We introduce a simple architecture based on similarity-distribution scores which we name Compositional Network generalization (CoRelNet) We find that simple architectural choices can outperform existing models in out-of-distribution generalizations.
  arXiv  Detail & Related papers  (2022-06-09T16:24:01Z)
• Prototypical Representation Learning for Relation Extraction [56.501332067073065]
  This paper aims to learn predictive, interpretable, and robust relation representations from distantly-labeled data. We learn prototypes for each relation from contextual information to best explore the intrinsic semantics of relations. Results on several relation learning tasks show that our model significantly outperforms the previous state-of-the-art relational models.
  arXiv  Detail & Related papers  (2021-03-22T08:11:43Z)

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.