研究如何让NER模型在嘈杂语音中新增隐私信息识别，发现标签机制可避免冲突。
Diagnosing Representation Dynamics in NER Model Extension
- 用增量学习诊断模型，区分语义与模式特征的独立性。
- 位置实体因与新隐私数据特征重叠而易出错，如邮政编码。
- 'O'标签分类器需解冻才能释放被误判的隐私模式，适合模型调试者。
将命名实体识别（NER）模型扩展至噪声语音数据中的新型个人身份信息（PII）实体是一项常见需求。我们发现，在标准语义实体（PER、LOC、ORG）和基于模式的新型PII（EMAIL、PHONE）上联合微调BERT模型，对原有类别影响极小。为此，我们探究这种“和平共存”现象，假设模型采用独立的语义与形态特征机制。通过增量学习设置作为诊断工具，我们测量了语义漂移，得到两个关键发现：第一，地点实体（LOC）因与新PII存在表示重叠（如邮政编码），具有独特脆弱性；第二，识别到‘反向O标签表示漂移’——模型初始将PII模式映射为‘O’标签，阻碍了新学习。只有解冻‘O’标签分类器，背景类才能自适应调整，从而‘释放’这些模式。本工作提供了对NER模型扩展的机理诊断，强调特征独立性、表示重叠及‘O’标签可塑性。研究基于Papernest收集的数据完成。
原文摘要 · Abstract (English)
Extending Named Entity Recognition (NER) models to new PII entities in noisy spoken-language data is a common need. We find that jointly fine-tuning a BERT model on standard semantic entities (PER, LOC, ORG) and new pattern-based PII (EMAIL, PHONE) results in minimal degradation for original classes. We investigate this "peaceful coexistence," hypothesizing that the model uses independent semantic vs. morphological feature mechanisms. Using an incremental learning setup as a diagnostic tool, we measure semantic drift and find two key insights. First, the LOC (location) entity is uniquely vulnerable due to a representation overlap with new PII, as it shares pattern-like features (e.g., postal codes). Second, we identify a "reverse O-tag representation drift." The model, initially trained to map PII patterns to 'O', blocks new learning. This is resolved only by unfreezing the 'O' tag's classifier, allowing the background class to adapt and "release" these patterns. This work provides a mechanistic diagnosis of NER model adaptation, highlighting feature independence, representation overlap, and 'O' tag plasticity. Work done based on data gathered by https://www.papernest.com
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。