Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.12702 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908920036982784 |
|---|---|
| author | Sun, Chaojie Cao, Bin Li, Tiantian Hou, Chenyu Li, Ruizhe Fan, Jing |
| author_facet | Sun, Chaojie Cao, Bin Li, Tiantian Hou, Chenyu Li, Ruizhe Fan, Jing |
| contents | With the rapid advancement of large language models (LLMs), growing efforts have been made on LLM-based table retrieval. However, existing studies typically focus on single-table query, and implement it by similarity matching after encoding the entire table. These methods usually result in low accuracy due to their coarse-grained encoding which incorporates much query-irrelated data, and are also inefficient when dealing with large tables, failing to fully utilize the reasoning capabilities of LLM. Further, multi-table query is under-explored in retrieval tasks. To this end, we propose a hierarchical multi-table query method based on LLM: Fine-Grained Multi-Table Retrieval FGTR, a new retrieval paradigm that employs a human-like reasoning strategy. Through hierarchical reasoning, FGTR first identifies relevant schema elements and then retrieves the corresponding cell contents, ultimately constructing a concise and accurate sub-table that aligns with the given query. To comprehensively evaluate the performance of FGTR, we construct two new benchmark datasets based on Spider and BIRD . Experimental results show that FGTR outperforms previous state-of-the-art methods, improving the F_2 metric by 18% on Spider and 21% on BIRD, demonstrating its effectiveness in enhancing fine-grained retrieval and its potential to improve end-to-end performance on table-based downstream tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_12702 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | FGTR: Fine-Grained Multi-Table Retrieval via Hierarchical LLM Reasoning Sun, Chaojie Cao, Bin Li, Tiantian Hou, Chenyu Li, Ruizhe Fan, Jing Information Retrieval Computation and Language Machine Learning With the rapid advancement of large language models (LLMs), growing efforts have been made on LLM-based table retrieval. However, existing studies typically focus on single-table query, and implement it by similarity matching after encoding the entire table. These methods usually result in low accuracy due to their coarse-grained encoding which incorporates much query-irrelated data, and are also inefficient when dealing with large tables, failing to fully utilize the reasoning capabilities of LLM. Further, multi-table query is under-explored in retrieval tasks. To this end, we propose a hierarchical multi-table query method based on LLM: Fine-Grained Multi-Table Retrieval FGTR, a new retrieval paradigm that employs a human-like reasoning strategy. Through hierarchical reasoning, FGTR first identifies relevant schema elements and then retrieves the corresponding cell contents, ultimately constructing a concise and accurate sub-table that aligns with the given query. To comprehensively evaluate the performance of FGTR, we construct two new benchmark datasets based on Spider and BIRD . Experimental results show that FGTR outperforms previous state-of-the-art methods, improving the F_2 metric by 18% on Spider and 21% on BIRD, demonstrating its effectiveness in enhancing fine-grained retrieval and its potential to improve end-to-end performance on table-based downstream tasks. |
| title | FGTR: Fine-Grained Multi-Table Retrieval via Hierarchical LLM Reasoning |
| topic | Information Retrieval Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2603.12702 |