Def-DTS uses a multi-step LLM prompt with utterance intent classification and a hard-coded deductive rule to segment dialogues, reporting state-of-the-art results on two of three benchmarks.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Def-DTS: Deductive Reasoning for Open-domain Dialogue Topic Segmentation
Def-DTS uses a multi-step LLM prompt with utterance intent classification and a hard-coded deductive rule to segment dialogues, reporting state-of-the-art results on two of three benchmarks.