Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

PRIMO: Progressive Induction for Multi-hop Open Rule Generation

Jianyu Liu, Sheng Bi, Guilin Qi

TL;DR

This work proposes a progressive multi-stage open rule generation method called PRIMO, which introduces ontology information during the rule generation stage to reduce ambiguity and improve rule accuracy, and employs reinforcement learning from human feedback to further optimize model.

Abstract

Open rule refer to the implication from premise atoms to hypothesis atoms, which captures various relations between instances in the real world. Injecting open rule knowledge into the machine helps to improve the performance of downstream tasks such as dialogue and relation extraction. Existing approaches focus on single-hop open rule generation, ignoring multi-hop scenarios, leading to logical inconsistencies between premise and hypothesis atoms, as well as semantic duplication of generated rule atoms. To address these issues, we propose a progressive multi-stage open rule generation method called PRIMO. We introduce ontology information during the rule generation stage to reduce ambiguity and improve rule accuracy. PRIMO constructs a multi-stage structure consisting of generation, extraction, and ranking modules to fully leverage the latent knowledge within the language model across multiple dimensions. Furthermore, we employ reinforcement learning from human feedback to further optimize model, enhancing the model's understanding of commonsense knowledge. Experiments show that compared to baseline models, PRIMO significantly improves rule quality and diversity while reducing the repetition rate of rule atoms.

PRIMO: Progressive Induction for Multi-hop Open Rule Generation

TL;DR

This work proposes a progressive multi-stage open rule generation method called PRIMO, which introduces ontology information during the rule generation stage to reduce ambiguity and improve rule accuracy, and employs reinforcement learning from human feedback to further optimize model.

Abstract

Open rule refer to the implication from premise atoms to hypothesis atoms, which captures various relations between instances in the real world. Injecting open rule knowledge into the machine helps to improve the performance of downstream tasks such as dialogue and relation extraction. Existing approaches focus on single-hop open rule generation, ignoring multi-hop scenarios, leading to logical inconsistencies between premise and hypothesis atoms, as well as semantic duplication of generated rule atoms. To address these issues, we propose a progressive multi-stage open rule generation method called PRIMO. We introduce ontology information during the rule generation stage to reduce ambiguity and improve rule accuracy. PRIMO constructs a multi-stage structure consisting of generation, extraction, and ranking modules to fully leverage the latent knowledge within the language model across multiple dimensions. Furthermore, we employ reinforcement learning from human feedback to further optimize model, enhancing the model's understanding of commonsense knowledge. Experiments show that compared to baseline models, PRIMO significantly improves rule quality and diversity while reducing the repetition rate of rule atoms.

Paper Structure

This paper contains 18 sections, 4 equations, 3 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Problem Definition
  4. PRIMO
  5. Generation
  6. Extraction
  7. Ranking
  8. Multi-hop Open Rule Generation
  9. RLHF
  10. Experiment
  11. Dataset
  12. Baselines
  13. Main Results
  14. Semantics Repetition
  15. The Length of Rule Chain
  16. ...and 3 more sections

Figures (3)

  • Figure 1: PRIMO consists of three modules. Generation module creates descriptions of hypothesis atoms based on the premise atoms. Extraction module extract atoms implied in text output of the generation stage. Rank module evaluates the plausibility of candidate hypothesis atoms.
  • Figure 2: We construct dataset and collect training corpus by ChatGPT with three step. Step1: use G_prompt to instruct ChatGPT to generate text that describes the relation between two entities. Step2: text from Step 1 is filled into E_prompt to extract hypothesis atoms. Step3: Ranking these hypothesis atoms through R_prompt.
  • Figure 3: Statistics of length of rule chains.