A new prompting strategy for multi-turn dialogues improves core information filtering by 32.6% and QA accuracy by 14.1% on HotpotQA while reducing inference time by 73.1% and tokens by 59.4%.
Constructing a multi-hop QA dataset for com- prehensive evaluation of reasoning steps
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A State-Update Prompting Strategy for Efficient and Robust Multi-turn Dialogue
A new prompting strategy for multi-turn dialogues improves core information filtering by 32.6% and QA accuracy by 14.1% on HotpotQA while reducing inference time by 73.1% and tokens by 59.4%.