mlsanityData to 14 Oct 2026

Solving Hard Analogy Questions with Relation Embedding Chains

arXiv 2310.12379 ↗First posted 18 Oct 2023cs.CL cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Modelling how concepts are related is a central topic in Lexical Semantics. A common strategy is to rely on knowledge graphs (KGs) such as ConceptNet, and to model the relation between two concepts as a set of paths. However, KGs are limited to a fixed set of relation types, and they are incomplete and often noisy. Another strategy is to distill relation embeddings from a fine-tuned language model. However, this is less suitable for words that are only indirectly related and it does not readily allow us to incorporate structured domain knowledge. In this paper, we aim to combine the best of both worlds. We model relations as paths but associate their edges with relation embeddings. The paths are obtained by first identifying suitable intermediate words and then selecting those words for which informative relation embeddings can be obtained. We empirically show that our proposed representations are useful for solving hard analogy questions.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 15 of its 45 references that are in the map. First posted on arXiv 18 Oct 2023.

Builds on

15 of 45 references.

References · 15

  1. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,372
  2. 2016ConceptNet 5.5: An Open Multilingual Graph of General Knowledgecited by 1,067
  3. 2021QA-GNN: Reasoning with Language Models and Knowledge Graphs for Question Answeringcited by 388
  4. 2019KagNet: Knowledge-Aware Graph Networks for Commonsense Reasoningcited by 216
  5. 2019Scene Graph Generation with External Knowledge and Image Reconstructioncited by 102
  6. 2022Subgraph Retrieval Enhanced Model for Multi-hop Knowledge Base Question Answeringcited by 93
  7. 2021BERT is to NLP what AlexNet is to CV: Can Pre-Trained Language Models Identify Analogies?cited by 56
  8. 2021JointLK: Joint Reasoning with Language Models and Knowledge Graphs for Commonsense Question Answeringcited by 40
  9. 2022Knowledge Base Question Answering by Case-based Reasoning over Subgraphscited by 31
  10. 2022E-KAR: A Benchmark for Rationalizing Natural Language Analogical Reasoningcited by 28
  11. 2022Life is a Circus and We are the Clowns: Automatically Finding Analogies between Situations and Processescited by 22
  12. 2023ANALOGYKB: Unlocking Analogical Reasoning of Language Models with A Million-scale Knowledge Basecited by 16
3 more
  1. 2021Distilling Relation Embeddings from Pre-trained Language Modelscited by 15
  2. 2013Domain and Function: A Dual-Space Model of Semantic Relations and Compositionscited by 13
  3. 2022Great Truths are Always Simple: A Rather Simple Knowledge Encoder for Enhancing the Commonsense Reasoning Capacity of Pre-Trained Modelscited by 9

Built on by

Not cited yet.