DISC: Plug-and-Play Decoding Intervention with Similarity of Characters for Chinese Spelling Check

2024-12-17

arXiv.org

AI Summary

In Chinese spelling correction (CSC), erroneous characters are highly phonetically or orthographically similar to their correct counterparts, posing two key challenges for existing methods relying on static confusion sets: difficulty in constructing comprehensive, high-quality sets and inability to model character-level substitution probabilities. This paper proposes DISC, a lightweight, plug-and-play decoding intervention module that operates solely during inference. DISC injects learnable phonetic–orthographic similarity signals—derived from joint phonetic encoding and structural character features—into the logits layer before Softmax, enabling model adaptation without retraining. Crucially, DISC is the first approach to formulate phonetic–orthographic similarity as a dynamic, decoding-stage intervention mechanism, thereby eliminating the need for explicit confusion set construction and probabilistic calibration. It supports zero-cost transfer across diverse CSC models. Evaluated on SIGHAN13/14/15 benchmarks, DISC consistently boosts performance of mainstream models, matching or surpassing state-of-the-art methods.

Technology Category

• Search and Optimization: Mixed Discrete/Continuous Search
• Reasoning under Uncertainty: Relational Probabilistic Models
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

One key characteristic of the Chinese spelling check (CSC) task is that incorrect characters are usually similar to the correct ones in either phonetics or glyph. To accommodate this, previous works usually leverage confusion sets, which suffer from two problems, i.e., difficulty in determining which character pairs to include and lack of probabilities to distinguish items in the set. In this paper, we propose a light-weight plug-and-play DISC (i.e., decoding intervention with similarity of characters) module for CSC models.DISC measures phonetic and glyph similarities between characters and incorporates this similarity information only during the inference phase. This method can be easily integrated into various existing CSC models, such as ReaLiSe, SCOPE, and ReLM, without additional training costs. Experiments on three CSC benchmarks demonstrate that our proposed method significantly improves model performance, approaching and even surpassing the current state-of-the-art models.

Problem

• Addresses Chinese spelling check (CSC) for phonetically or graphically similar characters
• Overcomes limitations of confusion sets in CSC models
• Enhances model performance without additional training costs

Innovation

• Plug-and-play module for Chinese spelling check
• Measures phonetic and glyph similarities
• Improves performance without training costs