Papers with synthetic query generation
It’s All Relative! – A Synthetic Query Generation Approach for Improving Zero-Shot Relevance Prediction (2024.findings-naacl)
Copied to clipboard
| Challenge: | Large language models generate synthetic query-document pairs by prompting with as few as 8 demonstrations. |
| Approach: | They propose to generate queries simultaneously for different labels by prompting with 8 demonstrations. |
| Outcome: | Extensive experimentation shows that synthetic queries generated in such a fashion improve performance. |
In-Context Reinforcement Learning with Retrieval-Augmented Generation for Text-to-SQL (2025.coling-main)
Copied to clipboard
| Challenge: | Existing methods of synthetic query generation generate mostly simple queries which might not be sufficiently representative of complex, real world queries. |
| Approach: | They propose to use large language models to fine tune query generation to produce complex queries that practitioners may pose during inference. |
| Outcome: | The proposed framework achieves 15-20% higher recall in database/table retrieval task compared to the existing state-of-the-art models for schema identification and upto 2% higher execution accuracy for SQL generation. |
Link, Synthesize, Retrieve: Universal Document Linking for Zero-Shot Information Retrieval (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for zero-shot learning are sparse, but have been used for dense retrieval (DR) . |
| Approach: | They propose a novel Universal Document Linking algorithm which links similar documents to enhance synthetic query generation across multiple datasets with different characteristics. |
| Outcome: | The proposed algorithm surpasses state-of-the-art methods in zero-shot cases. |