Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.29085 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911556267147264 |
|---|---|
| author | Li, Xingyu Wang, Rongguang Wang, Yuying Guo, Mengqing Li, Chenyang Sheng, Tao Ravi, Sujith Roth, Dan |
| author_facet | Li, Xingyu Wang, Rongguang Wang, Yuying Guo, Mengqing Li, Chenyang Sheng, Tao Ravi, Sujith Roth, Dan |
| contents | Large language models (LLMs) remain brittle on multi-hop question answering (MHQA), where answering requires combining evidence across documents through retrieval and reasoning. Iterative retrieval systems can fail by locking onto an early low-recall trajectory and amplifying downstream errors, while planning-only approaches may produce static query sets that cannot adapt when intermediate evidence changes. We propose \textbf{Planned Active Retrieval and Reasoning RAG (PAR$^2$-RAG)}, a two-stage framework that separates \emph{coverage} from \emph{commitment}. PAR$^2$-RAG first performs breadth-first anchoring to build a high-recall evidence frontier, then applies depth-first refinement with evidence sufficiency control in an iterative loop. Across four MHQA benchmarks, PAR$^2$-RAG consistently outperforms existing state-of-the-art baselines, compared with IRCoT, PAR$^2$-RAG achieves up to \textbf{23.5\%} higher accuracy, with retrieval gains of up to \textbf{10.5\%} in NDCG. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_29085 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | PAR$^2$-RAG: Planned Active Retrieval and Reasoning for Multi-Hop Question Answering Li, Xingyu Wang, Rongguang Wang, Yuying Guo, Mengqing Li, Chenyang Sheng, Tao Ravi, Sujith Roth, Dan Artificial Intelligence Large language models (LLMs) remain brittle on multi-hop question answering (MHQA), where answering requires combining evidence across documents through retrieval and reasoning. Iterative retrieval systems can fail by locking onto an early low-recall trajectory and amplifying downstream errors, while planning-only approaches may produce static query sets that cannot adapt when intermediate evidence changes. We propose \textbf{Planned Active Retrieval and Reasoning RAG (PAR$^2$-RAG)}, a two-stage framework that separates \emph{coverage} from \emph{commitment}. PAR$^2$-RAG first performs breadth-first anchoring to build a high-recall evidence frontier, then applies depth-first refinement with evidence sufficiency control in an iterative loop. Across four MHQA benchmarks, PAR$^2$-RAG consistently outperforms existing state-of-the-art baselines, compared with IRCoT, PAR$^2$-RAG achieves up to \textbf{23.5\%} higher accuracy, with retrieval gains of up to \textbf{10.5\%} in NDCG. |
| title | PAR$^2$-RAG: Planned Active Retrieval and Reasoning for Multi-Hop Question Answering |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2603.29085 |