A model fine-tuned on its own decomposition, retrieval, and critique traces improves multi-hop QA accuracy by 3 to 6 points on 2Wiki, HotpotQA, and MuSiQue.
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
-
RISE: Reasoning Enhancement via Iterative Self-Exploration in Multi-hop Question Answering
A model fine-tuned on its own decomposition, retrieval, and critique traces improves multi-hop QA accuracy by 3 to 6 points on 2Wiki, HotpotQA, and MuSiQue.