Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2502.05944 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866910820079763456 |
|---|---|
| author | Coleman, Jackson Lawrence, Isaiah Turner, Benjamin |
| author_facet | Coleman, Jackson Lawrence, Isaiah Turner, Benjamin |
| contents | Multi-source multi-hop question answering (QA) represents a challenging task in natural language processing due to the need for dynamic integration of heterogeneous knowledge sources and multi-step reasoning. Existing methods often suffer from cascading errors, insufficient handling of knowledge conflicts, and computational inefficiency. In this paper, we propose Adaptive Multi-source Knowledge-Oriented Reasoning (AMKOR), a generative framework that leverages large language models (LLMs) to dynamically fuse parametric and retrieved knowledge while exploring reasoning trajectories using probabilistic beam reasoning. AMKOR is further enhanced by a multi-granular learning strategy, optimizing both local reasoning steps and global answer accuracy. Experiments conducted on four widely-used multi-hop QA datasets, including HotpotQA and MuSiQue, demonstrate that AMKOR achieves state-of-the-art performance, significantly outperforming baseline methods on both reasoning accuracy and robustness. Additional analyses confirm its scalability, adaptability to noisy knowledge, and superior ability to handle complex multi-hop tasks. This work establishes a new benchmark for multi-source multi-hop QA by effectively combining reasoning quality and efficiency. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_05944 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Multi-granular Training Strategies for Robust Multi-hop Reasoning Over Noisy and Heterogeneous Knowledge Sources Coleman, Jackson Lawrence, Isaiah Turner, Benjamin Computation and Language Multi-source multi-hop question answering (QA) represents a challenging task in natural language processing due to the need for dynamic integration of heterogeneous knowledge sources and multi-step reasoning. Existing methods often suffer from cascading errors, insufficient handling of knowledge conflicts, and computational inefficiency. In this paper, we propose Adaptive Multi-source Knowledge-Oriented Reasoning (AMKOR), a generative framework that leverages large language models (LLMs) to dynamically fuse parametric and retrieved knowledge while exploring reasoning trajectories using probabilistic beam reasoning. AMKOR is further enhanced by a multi-granular learning strategy, optimizing both local reasoning steps and global answer accuracy. Experiments conducted on four widely-used multi-hop QA datasets, including HotpotQA and MuSiQue, demonstrate that AMKOR achieves state-of-the-art performance, significantly outperforming baseline methods on both reasoning accuracy and robustness. Additional analyses confirm its scalability, adaptability to noisy knowledge, and superior ability to handle complex multi-hop tasks. This work establishes a new benchmark for multi-source multi-hop QA by effectively combining reasoning quality and efficiency. |
| title | Multi-granular Training Strategies for Robust Multi-hop Reasoning Over Noisy and Heterogeneous Knowledge Sources |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2502.05944 |