(a) Average #candidates to link increase with different SQL difficulties. On our new dataset, RAT-SQL odel has more linking numbers (orange column) compared with the original dataset Spider (blue column). After applying our method (gray column), the linking number drops. (b) Average encoding length also increases with different SQL difficulties. More aggregate tables will increase the encoding length sharply. After applying our method, the situation is mitigated.
This site uses cookies. By continuing to use our website, you are agreeing to our privacy policy. No content on this site may be used to train artificial intelligence systems without permission in writing from the MIT Press.