AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

RAG-R1:Incentivizing the Search and Reasoning Capabilities of LLMs Through Multi-Query Parallelism

Zhiwen Tan, Jiaming Huang, Qintong Wu, Hongxuan Zhang, Chenyi Zhuang, Jinjie Gu

AAAI 2026 · Large language models › Retrieval and knowledge · approach: RAG robustness and evaluation · 18 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add format-learning SFT before RL (ablation)RAG-R1-sq · multi-hop QA averageEM: 0.318 -> 0.410Improved
add RL stage after SFT (ablation)RAG-R1-sq · multi-hop QA averageEM: 0.344 -> 0.410Improved
add RL data selection (filter unanswerable samples) (ablation)RAG-R1-sq · multi-hop QA averageEM: 0.370 -> 0.410Improved
add RL data selection (ablation)RAG-R1-mq · multi-hop QA averageEM: 0.411 -> 0.423Improved
add format-learning SFT before RL (ablation)RAG-R1-mq · multi-hop QA averageEM: 0.339 -> 0.423Improved
move from single-query to multi-query parallelismRAG-R1 · 7 QA benchmarks averageEM: 0.449 -> 0.457Improved
move from single-query to multi-query parallelismRAG-R1 · NQEM: 0.429 -> 0.423Hurt
move from single-query to multi-query parallelismRAG-R1 · 2WikiMultiHopQAEM: 0.520 -> 0.563Improved
move from single-query to multi-query parallelismRAG-R1 · inference timeseconds: 8.21 -> 7.42Improved
use multi-query instead of single-query in prompted Qwen2.5-72B-InstructSearch-R1-style agent · 2WikiMultiHopQAEM (%): 45.59 -> 50.45Improved
use multi-query instead of single-query in prompted Qwen2.5-72B-InstructSearch-R1-style agent · HotpotQAEM (%): 43.79 -> 44.33Improved
use RAG-R1-mq instead of R1-SearcherRL-based search agents · 7 QA benchmarks averageEM: 0.402 -> 0.457Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy