Using BERT probabilities to shrink the label list in LLM prompts improves intent recognition in multi-party dialogues and cuts latency, though gains vary by model and dataset.
In Findings of the Association for Computational Lin- guistics: EMNLP 2023, pages 3926–3939, Singapore
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
-
Intent Recognition and Out-of-Scope Detection using LLMs in Multi-party Conversations
Using BERT probabilities to shrink the label list in LLM prompts improves intent recognition in multi-party dialogues and cuts latency, though gains vary by model and dataset.