提出CobSeg模型，精准分割对话话题边界，提升语义连贯性识别能力。
CobSeg: Coherence Boundary Modeling for Dialogue Topic Segmentation
- 分离语义连贯性与词汇边界信号，通过方向性边界预测恢复两者
- 在5个基准上显著降低错误率，尤其在词汇线索明显时表现优异
- 无需大模型调用，适合实时对话系统部署，适合关注高效对话分析的开发者
对话话题分割在人机协作应用中至关重要，需识别异质的边界线索，包括话语边缘附近的词汇过渡和跨话语的语义断裂。现有话语模型常弱化局部词汇信号。本文提出CobSeg，一种多分支架构，将一致性层面的语义连续性与词汇边界转换分离，并通过方向性边界预测恢复二者。CobSeg进一步引入边界信息权重机制，强调高价值话语位置，并融合语料库导出的话题一致性提示，结合学习权重。在监督金标准训练和自动边界伪标签设置下评估，该模型在推理阶段无需大语言模型调用即可实现增强的边界预测。在五个基准上均取得改进：在金标准下，于VHF数据集上降低$P_k$ 0.7点、$W_d$ 0.6点，在DialSeg711达到$P_k$ 1.0；在伪标签设置下，于VHF降低$P_k$ 14.8点，于DialSeg711降低1.5点，于TIAGE降低1.1点，优于以往非大模型方法。
原文摘要 · Abstract (English)
Dialogue topic segmentation is critical in many human-AI collaborative applications which requires identifying heterogeneous boundary cues, including lexical transitions near utterance edges and semantic discontinuities across utterances. Existing utterance models often dilute these local lexical signals. We propose CobSeg, a novel multi-branch architecture that separates coherence-level semantic continuity from lexical boundary transitions and recovers both through directional boundary prediction. CobSeg further uses boundary informativeness weighting to emphasize high-utility utterance positions, and incorporates a corpus-derived topic coherence cue with learned combination weights. While CobSeg is evaluated as a compact trainable segmenter under supervised gold-boundary training and a pseudo-label setting with automatically induced boundaries, it performs enhanced boundary prediction without LLM calls during inference. Across five benchmarks, it improves $P_k$ and $W_d$ particularly when local lexical cues are prominent: under gold supervision, it reduces $P_k$ by 0.7 points and $W_d$ by 0.6 points on VHF, and reaches $P_k$ of 1.0 on DialSeg711; with induced boundaries, it reduces $P_k$ by 14.8 points on VHF, by 1.5 points on DialSeg711, and by 1.1 points on TIAGE, outperforming prior non-LLM approaches.
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。