Learning Complementary Action Modeling from Automotive Maintenance Instructions

2026-06-26

AI Summary

This work addresses the challenge that subtle variations in action phrases within automotive repair instructions can lead to semantically opposite procedural outcomes, necessitating the identification and generation of contextually consistent complementary instructions. The paper introduces a novel task termed “complementary action modeling,” which focuses on procedural opposition at the action-phrase level driven by minor lexical cues, distinguishing it from conventional approaches based on sentence similarity or paraphrasing. By integrating candidate matching with controlled Seq2Seq generation and validating results through multi-dimensional evaluation—including retrieval performance, lexical overlap, and human assessment—the proposed method effectively models complementary repair instructions. Experimental results demonstrate significant improvements over standard similarity-based modeling approaches in both semantic controllability and generation accuracy.

Abstract

A minute lexical variation can reverse the procedural meaning of an instruction even when the rest of the sentence remains unchanged. In automotive maintenance instructions, this pattern often appears when an action phrase turns an instruction into its procedural counterpart. The entities, modifiers, and surrounding context remain largely invariant, while the action phrase determines the procedural relation. We define this task as Complementary Action Modeling (CAM). Given a maintenance instruction, the goal is to identify or generate its procedural counterpart by modifying the action phrase while preserving the remaining sentence context. This task focuses on three aspects: distinguishing complementarity from surface similarity, controlling generation at the action-phrase level, and evaluating relational correctness using retrieval, overlap-based, and human evaluation. Using a German automotive maintenance dataset, we examine these questions through candidate matching and controlled Seq2Seq generation. The results show that complementary maintenance instructions are best modeled as procedural associations grounded in subtle lexical cues. They should therefore not be treated as ordinary cases of sentence similarity or synonym-based paraphrasing.

Problem

• Complementary Action Modeling
• procedural instructions
• action phrase
• lexical variation
• automotive maintenance

Innovation

• Complementary Action Modeling
• procedural counterpart
• action-phrase generation
• lexical cues
• controlled Seq2Seq