Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.15849 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917217449279488 |
|---|---|
| author | Chou, Tsung-Hsiang Yu, Chen-Jui Hsu, Shui-Hsiang Fan, Yao-Chung |
| author_facet | Chou, Tsung-Hsiang Yu, Chen-Jui Hsu, Shui-Hsiang Fan, Yao-Chung |
| contents | General-purpose embedding models have demonstrated strong performance in text retrieval but remain suboptimal for table retrieval, where highly structured content leads to semantic compression and query-table mismatch. Recent LLM-based retrieval augmentation methods mitigate this issue by generating synthetic queries, yet they often rely on heuristic partial-table selection and seldom leverage these synthetic queries as supervision to improve the embedding model. We introduce CGPT, a training framework that enhances table retrieval through LLM-generated supervision. CGPT constructs semantically diverse partial tables by clustering table instances using K-means and sampling across clusters to broaden semantic coverage. An LLM then generates synthetic queries for these partial tables, which are used in hard-negative contrastive fine-tuning to refine the embedding model. Experiments across four public benchmarks (MimoTable, OTTQA, FetaQA, and E2E-WTQ) show that CGPT consistently outperforms retrieval baselines, including QGpT, with an average R@1 improvement of 16.54 percent. In a unified multi-domain corpus setting, CGPT further demonstrates strong cross-domain generalization and remains effective even when using smaller LLMs for synthetic query generation. These results indicate that semantically guided partial-table construction, combined with contrastive training from LLM-generated supervision, provides an effective and scalable paradigm for large-scale table retrieval. Our code is available at https://github.com/yumeow0122/CGPT. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_15849 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | CGPT: Cluster-Guided Partial Tables with LLM-Generated Supervision for Table Retrieval Chou, Tsung-Hsiang Yu, Chen-Jui Hsu, Shui-Hsiang Fan, Yao-Chung Information Retrieval General-purpose embedding models have demonstrated strong performance in text retrieval but remain suboptimal for table retrieval, where highly structured content leads to semantic compression and query-table mismatch. Recent LLM-based retrieval augmentation methods mitigate this issue by generating synthetic queries, yet they often rely on heuristic partial-table selection and seldom leverage these synthetic queries as supervision to improve the embedding model. We introduce CGPT, a training framework that enhances table retrieval through LLM-generated supervision. CGPT constructs semantically diverse partial tables by clustering table instances using K-means and sampling across clusters to broaden semantic coverage. An LLM then generates synthetic queries for these partial tables, which are used in hard-negative contrastive fine-tuning to refine the embedding model. Experiments across four public benchmarks (MimoTable, OTTQA, FetaQA, and E2E-WTQ) show that CGPT consistently outperforms retrieval baselines, including QGpT, with an average R@1 improvement of 16.54 percent. In a unified multi-domain corpus setting, CGPT further demonstrates strong cross-domain generalization and remains effective even when using smaller LLMs for synthetic query generation. These results indicate that semantically guided partial-table construction, combined with contrastive training from LLM-generated supervision, provides an effective and scalable paradigm for large-scale table retrieval. Our code is available at https://github.com/yumeow0122/CGPT. |
| title | CGPT: Cluster-Guided Partial Tables with LLM-Generated Supervision for Table Retrieval |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2601.15849 |