TIRESRAG-R1 trains RAG models with sufficiency, thinking, and reflection rewards plus difficulty-aware reweighting, improving multi-hop QA accuracy.
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
-
From Sufficiency to Reflection: Reinforcement-Guided Thinking Quality in Retrieval-Augmented Reasoning for LLMs
TIRESRAG-R1 trains RAG models with sufficiency, thinking, and reflection rewards plus difficulty-aware reweighting, improving multi-hop QA accuracy.