CobSeg is a multi-branch architecture for dialogue topic segmentation that separates semantic continuity from lexical transitions, uses boundary informativeness weighting and corpus-derived cues, and reports metric gains on five benchmarks under both gold and induced boundary training.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
CobSeg: Coherence Boundary Modeling for Dialogue Topic Segmentation
CobSeg is a multi-branch architecture for dialogue topic segmentation that separates semantic continuity from lexical transitions, uses boundary informativeness weighting and corpus-derived cues, and reports metric gains on five benchmarks under both gold and induced boundary training.