Residual Skill Optimization for Text-to-SQL Ensembles
arXiv cs.CL·Jiongli Zhu, Haoquan Guan, Parjanya Prajakta Prashant, Nikki Lijing Kuang, Seyedeh Baharan Khatami, Canwen Xu, Xiaodong Yu, Yingyu Lin, Zhewei Yao, Yuxiong He, Babak Salimi
15h agoQuick Take
DivSkill-SQL optimizes Text-to-SQL ensembles by enhancing candidate diversity without model fine-tuning.
Key Points
- Improves Pass@K by targeting failures in current ensembles.
- Achieves up to +11.1 accuracy points on Snowflake.
- Reduces hallucinated schema references by up to 3x.
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CL
See more →arXiv cs.CL·Leyao Wang, Yanan He, Peng Chen, Asaf Yehudai, Yixin Liu, Rex Ying, Michal Shmueli-Scheuer, Arman Cohan
2d ago FeaturedOriginal
Time to REFLECT: Can We Trust LLM Judges for Evidence-based Research Agents?
AI Summary
The reliability of LLM judges for evaluating deep research agents is critically assessed using the REFLECT benchmark.