Unverified50% confidenceFactExact time
In text-to-SQL scenarios, schemagate first uses mathematical methods such as BM25, embedding and table-name matching to roughly filter candidate tables, then fuses and ranks them via RRF, and if the user provides an API key it calls an LLM to re-rank the top 20 candidate tables.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedRanked retrieval typically relies on mechanisms such as BM25, bi-encoder models, or cross-encoders to place the most relevant documents at the top.70% similarUnverifiedsqlglot's lineage() API requires schema information to work, and the author obtained the table structure in advance by connecting to Snowflake and executing DESCRIBE TABLE.69% similarVerifiedPrompt injection shares a structural analogy with SQL injection vulnerabilities: the latter exploits a database's inability to distinguish SQL code from data strings, while the former exploits an LLM's inability to distinguish meta-instructions from the content being processed.68% similarUnverifiedraggy adopts a hybrid database design, using both vector retrieval and BM25 keyword retrieval mechanisms simultaneously.68% similarUnverified该方案在SQL生成环节同时调用通义千问、DeepSeek、GLM三个大模型各自生成一条SQL，再由裁判机制从中选优67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1049510API
curl https://kongchang.com/api/v1/knowledge/claims/1049510MCP
get_claim(id=1049510)