Incomplete Utterance Rewriting with Editing Operation Guidance and Utterance Augmentation

2025-03-20

Conference on Empirical Methods in Natural Language Processing

AI Summary

To address the challenges of insufficient context focus, severe redundancy, and poor generalization due to scarce training data in incomplete utterance rewriting (IUR), this paper proposes EO-IUR, an edit-operation-guided multi-task learning framework. Our key contributions are: (1) a novel sequence-labeling-based supervision mechanism for edit operations, enabling token-level precise control over rewriting; (2) a two-dimensional utterance augmentation strategy integrating edit-operation-driven incomplete utterance augmentation and LLM-assisted historical dialogue augmentation; and (3) a dialogue-level heterogeneous lexical graph that models cross-turn, multi-granularity semantic relations. Evaluated on three benchmark datasets, EO-IUR significantly outperforms state-of-the-art methods, achieving consistent improvements in both rewriting accuracy and conciseness across open-domain and task-oriented dialogue scenarios.

Technology Category

• Natural Language Processing: Language Grounding & Multi-modal NLP
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

Although existing fashionable generation methods on Incomplete Utterance Rewriting (IUR) can generate coherent utterances, they often result in the inclusion of irrelevant and redundant tokens in rewritten utterances due to their inability to focus on critical tokens in dialogue context. Furthermore, the limited size of the training datasets also contributes to the insufficient training of the IUR model. To address the first issue, we propose a multi-task learning framework EO-IUR (Editing Operation-guided Incomplete Utterance Rewriting) that introduces the editing operation labels generated by sequence labeling module to guide generation model to focus on critical tokens. Furthermore, we introduce a token-level heterogeneous graph to represent dialogues. To address the second issue, we propose a two-dimensional utterance augmentation strategy, namely editing operation-based incomplete utterance augmentation and LLM-based historical utterance augmentation. The experimental results on three datasets demonstrate that our EO-IUR outperforms previous state-of-the-art (SOTA) baselines in both open-domain and task-oriented dialogue.

Problem

• Improves Incomplete Utterance Rewriting by focusing on critical tokens.
• Addresses dataset size limitations with a two-dimensional augmentation strategy.
• Introduces a multi-task learning framework for better dialogue representation.

Innovation

• Multi-task learning with editing operation guidance
• Token-level heterogeneous graph for dialogue representation
• Two-dimensional utterance augmentation strategy