DBScholar

Back to papers

BAT: Target-Instance-Free Data Preparation Synthesis via LLM-Driven Tree Search

Summary: BAT synthesizes executable data-preparation pipelines without target-table instances, supervision, or target access. An LLM-driven Monte Carlo tree search combines an action sandbox with execution-aware optimization, improving end-to-end pipeline accuracy by ≥8.74% over five baselines. (summarized by gpt-5.6-luna on Jul 26 2026)

Paper ID
7393
Venue
SIGMOD
Year
2026
Pagerank
5.093636e-05
Overall Rank
10,202 | 30.01%
DOI
10.1145/3802020

Incoming Non-self Citations Over Time

No non-self incoming citations found for this paper in this database.

Authors

BibTeX Citation

@inproceedings{ge_sigmod26,
        title = {{BAT: Target-Instance-Free Data Preparation Synthesis via LLM-Driven Tree Search}},
        author = {Ge, Congcong and Liu, Yachuan and Tang, Yixuan and Zhu, Yifan and Tu, Yaofeng and Gao, Yunjun},
        series = {{SIGMOD} '26},
        booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
        publisher = {Association for Computing Machinery},
        doi = {10.1145/3802020},
        url = {https://dl.acm.org/doi/10.1145/3802020},
        year = {2026}
}

Incoming Citations (Sorted by Pagerank)

Showing 0 of 0 citing papers.

Rank Citing Paper Year Venue Pagerank
Previous Page 1 / 1 Next

Outgoing Citations (Sorted by Pagerank)

Showing 12 of 12 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Previous Page 1 / 1 Next

Semantically Similar Papers