Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.21845 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916971279286272 |
|---|---|
| author | Zhang, Zichen Zhang, Kunlong Ruan, Hongwei Luo, Yiming |
| author_facet | Zhang, Zichen Zhang, Kunlong Ruan, Hongwei Luo, Yiming |
| contents | Transformer-based models have advanced the field of question answering, but multi-hop reasoning, where answers require combining evidence across multiple passages, remains difficult. This paper presents a comprehensive evaluation of retrieval strategies for multi-hop question answering within a retrieval-augmented generation framework. We compare cosine similarity, maximal marginal relevance, and a hybrid method that integrates dense embeddings with lexical overlap and re-ranking. To further improve retrieval, we adapt the EfficientRAG pipeline for query optimization, introducing token labeling and iterative refinement while maintaining efficiency. Experiments on the HotpotQA dataset show that the hybrid approach substantially outperforms baseline methods, achieving a relative improvement of 50 percent in exact match and 47 percent in F1 score compared to cosine similarity. Error analysis reveals that hybrid retrieval improves entity recall and evidence complementarity, while remaining limited in handling distractors and temporal reasoning. Overall, the results suggest that hybrid retrieval-augmented generation provides a practical zero-shot solution for multi-hop question answering, balancing accuracy, efficiency, and interpretability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_21845 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Comprehensive Evaluation of Transformer-Based Question Answering Models and RAG-Enhanced Design Zhang, Zichen Zhang, Kunlong Ruan, Hongwei Luo, Yiming Computer Vision and Pattern Recognition Transformer-based models have advanced the field of question answering, but multi-hop reasoning, where answers require combining evidence across multiple passages, remains difficult. This paper presents a comprehensive evaluation of retrieval strategies for multi-hop question answering within a retrieval-augmented generation framework. We compare cosine similarity, maximal marginal relevance, and a hybrid method that integrates dense embeddings with lexical overlap and re-ranking. To further improve retrieval, we adapt the EfficientRAG pipeline for query optimization, introducing token labeling and iterative refinement while maintaining efficiency. Experiments on the HotpotQA dataset show that the hybrid approach substantially outperforms baseline methods, achieving a relative improvement of 50 percent in exact match and 47 percent in F1 score compared to cosine similarity. Error analysis reveals that hybrid retrieval improves entity recall and evidence complementarity, while remaining limited in handling distractors and temporal reasoning. Overall, the results suggest that hybrid retrieval-augmented generation provides a practical zero-shot solution for multi-hop question answering, balancing accuracy, efficiency, and interpretability. |
| title | A Comprehensive Evaluation of Transformer-Based Question Answering Models and RAG-Enhanced Design |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2509.21845 |